The lives of rural women and girlsWhat does an inclusive agricultural transformation that empowers women look like?Louise Fox, Steve Wiggins, Eva Ludi and Anna Mdee

October 2018

Report

Readers are encouraged to reproduce material for their own publications, as long as they are not being sold commercially. ODI requests due acknowledgement and a copy of the publication. For online use, we ask readers to link to the original resource on the ODI website. The views presented in this paper are those of the author(s) and do not necessarily represent the views of ODI or our partners.

This work is licensed under CC BY-NC-ND 4.0.

Cover photo: A mother cooks food for her family, 2011 © UNICEF Ethiopia.

3

Acknowledgements

Many thanks to those who helped us compile this report. Caroline Harper provided guidance and overall commentary on the project. Likie Negussie interviewed subjects and collated literature in Addis Ababa; while Harriet Hoffler and Jassi Sandhar helped review the Ethiopian literature. Field work in Tanzania included Anasia Maleko gathering policy insights in Dar, and Gasper Mdee interviewing people in villages. For Egypt, valuable comments and insights were provided by Guy Jobbins and Dina Mansour-Ilie. For Peru, Álvaro Espinoza interviewed key informants in Lima. Sophie Bridonnneau, Ivor Jones and Sharada Keats helped analyse data and review literature. Ore Kolade and Tom Clark helped administer the studies.

Vicky Wilde and Chiara Kovarik from the Bill and Melinda Gates Foundation provided many helpful suggestions and literature. Twice we went to Seattle and benefitted from the questions and comments of staff there.

The paper was reviewed in detail by Clare Bishop who provided valuable recommendations and insights.

Deb Eades edited the copy. The authors, however, take responsibility for this report, and any errors and omissions it contains.

The report does not necessarily reflect the views of either the Overseas Development Institute or the Bill and Melinda Gates Foundation.

4

Contents

Acknowledgements 3

List of boxes, tables and figures 5

Acronyms 10

Executive summary 11

1 Overview 16

1.1 Motivation: improving the lives of rural women and girls 16

1.2 Aims, approach and methods 16

1.3 Findings 17

1.4 Ethiopia and Tanzania 24

1.5 Conclusions and discussion 25

2 Introduction 29

2.1 Motivation: improving the lives of rural women and girls 29

2.2 Aims of study, questions addressed 29

2.3 Approach and methods 30

3 Agricultural development and the lives of rural women and girls: a framework 31

3.1 Introduction to the framework 31

3.2 National structural transformation 34

3.3 Agricultural and rural transformation 36

3.4 Transformation of rural households 43

3.5 Gendered aspects of economic, agricultural, and rural transformations 45

3.6 Conclusion 50

4 Changing lives of rural women and girls over the longer run: Egypt, Peru and Thailand 53

4.1 Introduction 53

4.2 Thailand: an imperfect exemplar 56

4.3 Peru: a delayed start 70

4.4 Egypt: intractable conundrums? 81

4.5 Key messages from change in the long run 92

5

5 Changing lives of rural women and girls in contemporary Africa: Ethiopia and Tanzania 95

5.1 Introduction 95

5.2 Ethiopia 95

5.3 Tanzania 121

5.4 Conclusions for Tanzania 136

5.5 Conclusions. Insights from Ethiopia and Tanzania 136

6 Conclusions 140

6.1 Lessons from the country reviews 140

6.2 Looking forward. How can agricultural development help rural women and girls? 142

References 145

List of boxes, tables and figures

Boxes

Box 1 Timmer’s stages of agricultural development 38

Box 2 The importance of land tenure for growth with equity 42

Box 3 Thailand’s highly successful family planning program 59

Box 4 Transforming the Northeast from backwater to export powerhouse 62

Box 5 From structural transformation to individual blame: making sense of change in contemporary Asia 69

Box 6 Peru’s radical land reform: changing structures but not productivity 76

Box 7 Higher aspirations for rural girls 80

Box 8 A cautionary note on data 105

Box 9 Insights into agricultural growth from village and district studies 128

6

Tables

Table 1 How opportunities for women and girls expand with agricultural transformation 51

Table 2 Key statistics for development of Egypt, Peru and Thailand, 1965, 1990 and 2015 54

Table 3 Estimated annual increase in cereal yields, different sources, since 1997 101

Table 4 National FGM/C prevalence data disaggregated by age cohorts 2000–2016 119

Table 5 Summarising changes in four dimensions of rural women’s lives in the five countries 143

Figures

Figure 1 Inclusive agricultural transformation: levels of analysis 17

Figure 2 Economic growth, Thailand, 1961–2015 18

Figure 3 Fertility, Thailand, 1961–2014 19

Figure 4 Use of contraception, Thailand, 1970–2012 19

Figure 5 Agricultural productivity, Thailand, 1961–2013 20

Figure 6 Poverty, Thailand, 1988–2013 20

Figure 7 Women’s employment, Thailand, 1980–2013 20

Figure 8 Economic growth, Peru 1960–2015 22

Figure 9 Poverty and inequality, Peru, 1985–2015 22

Figure 10 Economic growth, Egypt, 1967–2015 24

Figure 11 Fertility, Egypt, 1988–2014 24

Figure 12 Economic growth, Ethiopia, 1981–2015 26

Figure 13 Economic growth, Tanzania, 1988–2015 26

Figure 14 Achieving Inclusive Rural Transformation: Levels of Analysis 34

Figure 15 Stages of agricultural development, after Timmer 1988 38

Figure 16 Economic growth, Thailand, 1961–2015 57

Figure 17 Structure of production, Thailand, 1961–2014 57

Figure 18 Population growth, Thailand, 1961–2015 58

Figure 19 Fertility, Thailand, 1961–2014 58

7

Figure 20 Women’s use of contraception, Thailand, 1970–2012 58

Figure 21 Agricultural growth, Thailand, 1961–2013, three year moving average 60

Figure 22 Agricultural productivity, Thailand, 1961–2013 60

Figure 23 Agricultural exports, Thailand, 1961–2013 61

Figure 24 Map of Northeast Thailand 63

Figure 25 Employment by sector, Thailand, 1960–2011 64

Figure 26 Poverty, Thailand, 1988–2013 65

Figure 27 Income inequality, Thailand, 1981–2013 65

Figure 28 Women’s jobs, Thailand, 1980–2013 66

Figure 29 Women’s wages, Thailand, 1991–2005 66

Figure 30 Life expectancy, Thailand, 1960–2014 67

Figure 32 School enrolment in Thailand, 1971–2012 67

Figure 31 Under-five mortality, Thailand, 1960–2014 67

Figure 33 Economic growth, Peru, 1960–2015 72

Figure 34 Structural change, Peru, 1960–2015 72

Figure 35 Population growth, Peru, 1960–2014 73

Figure 36 Fertility rate, Peru, 1986–2014 73

Figure 37 Use of contraception by adult married women aged 15–49, Peru, 1986 74

Figure 38 Agricultural growth, Peru, 1961–2013 74

Figure 39 Agricultural area, labor, and productivity, Peru, 1961–2013 75

Figure 40 Poverty and inequality, Peru, 1985–2015 77

Figure 41 Growth in public investment in rural areas as a percentage of GDP, 2002–2012 77

Figure 42 Under-five mortality, Peru, 1986–2014 78

Figure 43 Rural women’s education, Peru, 1992–2014 79

Figure 44 Reported spousal violence, Peru, 2004–2014 81

Figure 45 Economic growth, Egypt, 1967–2015 83

8

Figure 46 Structural transformation, Egypt, 1966–2015 83

Figure 47 Population growth, Egypt, 1961–2015 84

Figure 48 Fertility, Egypt, 1988–2014 84

Figure 49 Married women using contraception, Egypt, 1988–2014 84

Figure 50 Agricultural output, Egypt, 1962–2013 85

Figure 51 Agricultural area, labor and productivity, Egypt, 1961–2013 86

Figure 52 Mortality of children under five years, Egypt, 1988–2014 89

Figure 53 Rural women’s education, Egypt 1992–2014 89

Figure 54 Economic growth, Ethiopia, 1981–2015 97

Figure 55 Contributions to growth, Ethiopia, 2002–2012 97

Figure 56 Sector shares of GDP, Ethiopia, 1981–2015 97

Figure 57 Shares of employment, Ethiopia, 1994–2013 98

Figure 58 Population growth, Ethiopia, 1981–2015 98

Figure 59 Net Rate of Assistance to Agriculture, Ethiopia, 1981–2005 99

Figure 60 Agricultural growth, Ethiopia, 1962–2012 100

Figure 61 Increases in production, area and yield, main crops, Ethiopia, 1993/95 to 2012/14 100

Figure 62 Agricultural trade, Ethiopia, 1961–2013 100

Figure 63 Land and Labor Productivity, Ethiopia, 1993–2014 101

Figure 64 Travel time to reach a city of more than 50,000 persons, 1996/97 and 2010/11 103

Figure 65 Poverty headcount ratio at US$0.70/day, $1.25/day and $3.10/day 1995–2010 104

Figure 66 Poverty headcount ratio using the national poverty line in rural and urban areas 1995–2010 104

Figure 67 Consumption growth for the bottom 10%, bottom 40% and top 60% 1996–2011 106

Figure 68 Rural and urban multidimensional poverty 2000–2011 106

Figure 69 Average real wage from paid employment per sector and for women and men 107

Figure 70 Drinking water sources, rural Ethiopia, 1990 and 2015 110

Figure 71 Net attendance ratio, primary schools, Ethiopia, 2000–2011 112

9

Figure 72 Net attendance, secondary school, Ethiopia, 2000–2011 112

Figure 73 Human resources and health infrastructure, Ethiopia, 2001–2013 113

Figure 74 Coverage of key reproductive, maternal, new-born and child health outcomes, Ethiopia,

2000–2014 113

Figure 75 Maternal mortality ratio (MMR) with confidence intervals, Ethiopia, for the 7 years preceding

2000, 2005, 2011, and 2016 113

Figure 76 Total fertility rate, Ethiopia, 1990–2014 114

Figure 77 Women’s attitudes towards wife beating in rural areas, Ethiopia, 2000–2011 118

Figure 78 Economic growth, Tanzania, 1988–2015 122

Figure 79 Share of GDP by sector, Tanzania, 1990–2015 123

Figure 80 Improvement in Pillar Indices, Tanzania, 2012–2015 123

Figure 81 Population change, Tanzania, 1985–2015 124

Figure 82 Net rate of assistance to agriculture, Tanzania, 1976–2004 Population change, Tanzania,

1985–2015 124

Figure 83 Agricultural production, Tanzania, 1962–2011 126

Figure 84 Agricultural trade, Tanzania, 1961–2013 126

Figure 85 Arable land, agricultural labor and productivity, Tanzania, 1961–2011 127

Figure 86 Employment structure and growth in non-farm activity 129

Figure 87 Household sources of income 129

Figure 88 Number of employed people with secondary activity by type 131

Figure 89 Sector of activity of rural household enterprises, 2014 134

Figure 90 Structure of employment, main activity, by gender and area 134

Figure 91 Time spent on productive activities and household chores, Tanzania, 2006 and 2014 135

Figure 92 Net primary and secondary attendance, Tanzania, 2004/05 to 2015/16: girls are doing well 136

Figure 93 Under-five mortality, Tanzania, 1992–2016 136

Figure 94 Access to water, rural area, Tanzania, 1992 to 2016 136

Figure 95 Fertility, by region, Tanzania, 2015/16 137

10

Acronyms

3YMA Three-year moving average

ADLI Agricultural development-led industrialisation (Ethiopia)

AMC Agricultural Marketing Corporation, Ethiopia

ATA Agricultural Transformation Agency, Ethiopia

DHS Demographic and Health Survey

EDHS Ethiopia Demographic and Health Survey

FDI Foreign direct investment

FGM Female genital mutilation

GBV Gender-based violence

HE Household enterprise

IFPRI International Food Policy Research Institute

k Thousand

km Kilometer

LFS Labor Force Survey

MFI Micro-finance institution

MPI Multi-dimensional poverty index

One-to-fives Ethiopia: One-to-five networks are groups of six households living in the same neighborhood: the leader will report to government agencies

OPHI Oxford Poverty and Human Development Initiative, Oxford University

PPP Purchasing Power Parity

PSNP Productive Safety Net Program (Ethiopia)

RNFE Rural non-farm economy

RuSACCO Rural Savings and Credit Cooperatives (Ethiopia)

SME Small and medium enterprises

SNNPR Southern Nations, Nationalities and Peoples’ Region, Ethiopia

TASAF Tanzania Social Action Fund

tonne Metric tonne

VAW Violence against women

WDI World Development Indicators: World Bank database

11

Executive summary

Introduction

Across the global South, most rural women and girls are disadvantaged. Compared to men and boys they receive less formal education, some may get less health care, they have fewer opportunities to work outside the household and when they do, they are often paid less and treated worse than men. Most rural women are constrained by social norms that define them primarily as wives, mothers and confined to the domestic sphere, where men do less than their fair share of household chores. Women are typically expected to be subservient to men. At worst, they are subject to emotional and physical abuse by men.

Not only is this unjust, but also it means the full potential of rural women and girls – as people, workers, citizens, leaders – is not realised, to the detriment of their households and families, their rural communities and indeed their nations. When rural mothers lack the basics of life – food, income, health care – their children are at risk. Gender inequity thus threatens future generations.

Questions and approach

This study addresses two questions from the Bill and Melinda Gates Foundation:

• What does an inclusive agricultural transformation that empowers women look like?

• What are the implications for: (a) policy; (b) infrastructure; (c) technology; and (d) behavior change?

In addressing these questions, the study sets agricultural development within the wider context of economic growth and transformation.

Changes to the lives of rural women and girls have been framed within a in a series of development transformations – at national level, in agriculture and rural areas, in households and finally for individuals. The framework also sees agricultural development as progressing through stages – early, middle and late, which reflect the declining relative importance of on farm agriculture activities as national economies develop and diversify away from a dependence on agriculture.

Drivers of agricultural transformation may serve to stimulate or support inclusion of rural women and girls, or to prevent and reduce it. Whether such drivers help inclusion is an empirical question for specific cases.

Investigating how past transformations became inclusive, three cases of long-term rural economic transformation since the 1960s were studied: Egypt, Peru and Thailand. All three countries experienced economic growth, urbanisation, and a marked shift in their economic structure as agriculture has declined relative to industry and services. Available statistics were collected, and existing literature reviewed, to provide material for interpretive studies that set out the main changes in national and rural economies, and what has changed for rural women and girls.

In addition, two contemporary cases were examined: Ethiopia and Tanzania, both low-income countries that are still largely rural and agrarian, but with growing economies and transformations in progress. For both countries, the study relied mainly on existing statistics and literature, complemented by an exploration of policy for women’s empowerment. In addition, for Tanzania, a qualitative study of change for rural women and girls was undertaken in two contrasting areas of the country.

12

Findings

Of three countries studied over the longer run, Thailand is an exemplar of what is possible through growth and transformation. Since the early 1960s, when policy changed from industrialisation led by the state to that led by private firms, the Thai economy has grown rapidly. By 2015, GDP per person was almost ten times higher than in 1961, in real terms.

Growth was encouraged by policies and macroeconomic management that encouraged private investment; by public investments in roads, irrigation schemes, health and education; and, during the 1960s and 1970s, by aid from the United States that included funds to construct roads deep into rural areas.

Agriculture grew ahead of population growth, not only feeding a growing population, but also generating net exports. Agricultural development was largely extensive before the 1990s: new land, much of it taken from the forests, was brought into cultivation. Since 1990, agriculture has grown through increased land and labour productivity.

In rural areas, the non-farm economy expanded through new businesses servicing agriculture, through expanded production of goods and services for farmers with increased incomes, and some decentralised manufacturing. Growth of the urban economy created opportunities for young people from rural areas, including women, to migrate to work in retailing, domestic service, construction and assembly plants.

Meanwhile, owing to a successful family planning programme that covered rural areas, fertility rates fell from more than six per woman on average to under two by 2015. Population growth slowed notably.

The combination of a slowing growth of the labour supply, and many more employment opportunities, led to rising rural wages: farm wages tripled in real terms between 1985 and 2003.

The lives of rural women and girls have been profoundly affected by these economic and demographic changes. Diversification of the rural economy created new employment options for women, either as migrants or as workers in rural enterprises.

Rural women’s work has been eased by reduced fertility and hence fewer children to care for, and by electricity and running water in homes – even if women still do much more domestic work than men. Public investments in health and education mean that rural women and girls are healthier and better educated than their mothers and grandmothers were.

While women have faced gender norms that ascribe women a secondary role in a society where men are expected to lead and take key decisions, those norms have not prevented women taking up employment and business opportunities. Above all, young women have not been prevented from migrating.

Thailand shows how much progress for rural women can be made when broad-based growth takes place, along with public spending on rural public services, in a context where gender norms, while favouring men, do not block women’s progress. It also shows how beneficial a rapid demographic transition can be for rural transformation and women’s lives.

Peru dissipated its earlier development through losing almost 25 years of economic growth between the late 1960s and early 1990s, owing to the problems of land reform, unwise macroeconomic policy and civil war in parts of the Andean highlands.

Since the early 1990s, Peru’s economy has grown rapidly, initially on a narrow base of mining, and exports from commercial farming along the coast. In the 1990s that meant too few jobs were created, and poverty and inequality rose. But in the 2000s a combination of slowing population growth and job creation, notably in export agriculture, raised employment and wages. This, plus public spending on roads, irrigation, health, education and cash transfers in low-income areas, predominantly the highlands, led to major reductions in poverty and inequality since the early 2000s.

The material conditions of life for rural women have improved substantially: more economic opportunities, better health and education. Indeed, recent life histories indicate a sea-change in the perspectives of rural parents for their daughters. Seeing new off-farm

13

opportunities, they recognise that secondary schooling is critical for girls’ prospects.

As in Thailand, although the material conditions of rural women’s lives have improved, changes to gender norms and status have been slower.

Egypt provides a further contrast. Economic and agricultural growth have taken place, but inclusive growth has been hampered, with consequences for rural women who have made less progress than in the other two countries. Egypt’s growth model has not created enough non-farm employment to absorb the young and growing labour force, especially since the mid-1990s. Despite the economic liberalisation that might have attracted foreign investors, Egypt has not seen the growth of assembly plants and light engineering in the ways that so many Asian emerging economies have. Moreover, many of the jobs that have been created in Egypt have been regarded as male jobs – even in tourism - providing few opportunities for women.

Above all, Egypt is going through a prolonged, and possibly even stalled, demographic transition. Fertility rates remain stubbornly high. Hence, the labour force has grown faster than employment in most years since the mid-1990s. Not only has this condemned youth to a difficult search for employment that is often informal and ill paid; it has disproportionately affected women, especially rural women, who are less educated and less mobile than their urban counterparts.

Rural girls thus face the prospect that their future lives will be restricted to the domestic domain where they will be expected to do much of the farm work, bear and care for children, and be responsible for domestic chores. And to do so with few firm rights to property and under the threat of violence from husbands.

That said, poverty and deprivation are still the major determinants of life opportunities and options in Egypt, more so than gender. The search for gender equity goes hand in hand with poverty reduction.

Ethiopia and Tanzania are low-income countries, still predominantly agrarian. Both share a checkered economic past as their economies faltered in the 1970s and 1980s owing to failed state-led economic development. Since

then, however, both countries have reduced state intervention in the economy, and encouraged private investment, which has contributed to renewed economic growth and major improvements in living standards.

The lives of rural women and girls have largely improved since the 1990s. Economic growth and structural transformation, with increases in agricultural productivity – no matter how modest and patchy – mean that returns to rural labour have risen. Moreover, the growth of the rural non-farm economy and improved connections from rural areas to towns and cities has widened opportunities for rural labour.

Public provision of education, health care and clean water means that rural people, especially women and girls, have not only experienced significant improvements in their welfare; but also, that they are better able to take advantage of new opportunities created by growth and transformation.

Fertility has fallen in both countries, even if only slightly in Tanzania, reducing the amount of child care that rural women have to provide. In some rural areas, improved water supplies have cut the time taken to draw water, a task that generally falls to rural women and girls.

Norms that see a woman’s place as primarily a wife and mother, assign rights to property first and foremost to men, and tolerate worse treatment of women than men – including domestic violence – do not seem to have changed as much as material conditions have.

Progress for rural women, then, has been variable. Depending on which of the different dimensions of women’s empowerment are examined, it is possible to see promising progress; or to see stasis punctuated by occasional, but limited, improvements in material conditions.

Further progress for rural women in Ethiopia and Tanzania will largely depend on three conditions:

• Growth of the urban economy, manufacturing and the rural non-farm economy. Most of the above-average productivity, and hence higher wages, will be in these sectors.

• Family planning which not only leads to lower fertility, slowing the labour supply that

14

eventually will mean significant increases in wages and earnings but also contributes to empowering rural women.

• Girls’ secondary schooling, which supports women’s empowerment, by giving young women the qualifications that most formal employers seek and expect; by making early marriage less likely; and raising young women’s status.

Conclusions and discussion

Four main findings emerge from this study. One, broad-based development not only

creates many of the additional employment opportunities needed to meet the rapid expansion in the labour force when countries undergo a demographic transition; but also, jobs where average labour productivity rises as well.

Agricultural development is part and parcel of such growth and transformation. It is not just agriculture that counts, however: the growth of the rural non-farm economy (RNFE) also matters. Non-farm work grows either through activities that are linked to agriculture either directly in the supply chains up and downstream of the farm, or indirectly through spending by farm households. Public spending on rural infrastructure and services also creates rural jobs. Migration from rural to urban areas can raise also increase earnings.

The generation of both rural employment and migration options seems especially important for rural women. Within agriculture, women tend to farm crops and livestock at low labour productivity; are deprived of access to inputs and technical innovations; or both. Compared to this work, a job in the RNFE or migrating to the city can pay much more. Moreover, while working within the farm household, a woman may earn very little income in cash, or see her husband appropriate it. When working for wages or operating a small enterprise, a woman is much more likely to control her income. And with that control can often come greater status and self-esteem.

A simple proposition emerges from these cases: getting out of agriculture matters more for rural women than for rural men.

Two, rural people can only take up better economic opportunities if they are healthy, literate and numerate. Hence public investments in rural education, health, and clean water and sanitation make a real difference. Primary health services, water and sanitation, have done much to reduce disease: the decline in under-five mortality in all five countries, taking this a proxy for overall rural health, has been remarkable. In a generation, the pain of losing an infant has gone from being commonplace to a sad rarity.

Schooling in rural areas has increased notably in all five countries: for the middle-income countries, high rates of enrolment in secondary school are common, with little difference between girls and boys.

Three, one service matters more than expected: family planning. When family planning is readily accessible to rural women – which usually means services in the village, preferably on the doorstep, and provided by people they trust, in an overall supportive cultural environment, contraception is often rapidly and widely used. Not only does this put women in control of their fertility, but also it leads to fewer births, and – with a delay of a decade or two – slower growth of the labour force. Once rural populations cease to grow rapidly, labour shortages usually emerge, investment to raise labour productivity is encouraged, and wages for unskilled rural work rise.

Four, gender norms count. They are not be set in stone, and are often reinterpreted when the economic advantage is obvious – as in rural Thailand when it became clear in the 1970s that young rural women could earn valuable incomes if they migrated – yet nevertheless norms can either facilitate or block these processes. Mobility stands out: rural women have far more options when it is acceptable for them to travel independently, to visit market centers, to commute for jobs, or to migrate to the city. Of course, even in societies where women’s rights are firmly established in law, women who travel or migrate may still incur risks in personal safety and poor treatment at work. The point, however, is to reduce those risks, rather than prevent movement.

Finally, even if some of the strongest drivers of change for rural women and girls are not

15

specifically agricultural, this does not mean that the considerable interest in women in agriculture that has been awakened in the last ten or so years has been in vain. Far from it. Women farmers are disadvantaged and much can be done to remedy this.

Rural women’s rights – to land and property, to public services, to fair and decent treatment at work, to protection from violence – need strengthening. All these rights support women in rural areas, and when they seek to leave rural areas, as migrants.

16

1 Overview

1.1 Motivation: improving the lives of rural women and girls

Across the global South, most rural women and girls are disadvantaged. Compared to men and boys they receive less formal education, some may get less health care, they have fewer opportunities to work outside the household and when they do, they are often paid less and treated worse than men. Most rural women are constrained by social norms that define them primarily as wives, mothers and confined to the domestic sphere, where men do less than their fair share of household chores. Women are typically expected to be subservient to men. At worst, they are subject to emotional and physical abuse from men.

Not only is this unjust, but also it means the full potential of rural women and girls – as people, workers, citizens, leaders – is not realised, to the detriment of their households and families, their rural communities and indeed their nations. When rural mothers lack the basics of life – food, income, health care – their children are at risk. Gender inequity thus threatens future generations.

While the current reality is unsatisfactory, the lives of rural women and girls are changing. In many developing countries, it is much more common today for rural girls to go to school, for pregnant women to receive ante-natal care, for women to take on paid jobs and run businesses, and to migrate to towns and cities for better opportunities, than was even imaginable for their grandmothers.

The overall economic, political, and social development of these countries over the last 50 years has provided the means, and in some cases, the drivers, of these changes. Many formerly low-income countries have gone from being largely agrarian and rural to industrial and urban, and in the process, have seen incomes rise and living standards improve. They have also undergone a

demographic transition, from high to low rates of birth and death, slowing population growth.

Rural areas and agriculture may be relatively less important economically than they were at early stages of economic development, but it is in the rural areas that some of the largest development deficits can be still be seen: in poverty, hunger, low education, and poor health. Against this backdrop, this report looks at how the lives of rural women and girls can be improved through rural transformation.

1.2 Aims, approach and methods

Two questions were set for this study by the Bill and Melinda Gates Foundation:

• What does an inclusive agricultural transformation that empowers women look like?

• What are the implications for: (a) policy; (b) infrastructure; (c) technology; and (d) behavior change?

Rural women’s disadvantage in economic activities is especially marked in agriculture, as women farmers struggle with insecure access to land, poorer supplies of inputs, non-existent credit, and less technical knowledge than men farmers (FAO et al., 2010; FAO, 2011; World Bank and ONE, 2014).

This study, however, has not been limited to questions of agricultural development. Rather, agricultural development has been set within the wider context of economic growth and transformation.

The approach taken first sets out a framework to consider how economic development at national, rural, household and individual levels affect the lives of rural women and girls.

Then, using this framework, three cases of long-term rural economic transformation

17

since the 1960s were studied: Egypt, Peru and Thailand. All three countries have seen economic growth, urbanisation, and a marked shift in their economic structure as agriculture has declined relative to industry and services. Available statistics were collected and existing literature reviewed, to provide the material for interpretive studies that set out the main changes in national and rural economies, and what has changed for rural women and girls.

Finally, two contemporary cases were examined: Ethiopia and Tanzania, both low-income countries that are still largely rural and agrarian, but with growing economies and transformations in progress. For both countries, the study relied mainly on existing statistics and literature, complemented by an exploration of policy for women’s empowerment. In addition, for Tanzania, a qualitative study of change for rural women and girls was undertaken in two contrasting areas of the country.

1.3 Findings

1.3.1 Framing changes in the lives of rural women and girls Rural women’s lives are influenced by processes operating at the level of the overall economy, the rural economy, and among and within rural households (Figure 1).

Within transformation processes at each level are factors driving improved outcomes. These drivers may be initial conditions, or events external to the level of analysis – such as favourable terms of trade, investment climate, the availability of natural resources, etc. Or they may be policies and programmes, such as building cost-effective rural roads or land-titling initiatives that ensure that titles are jointly held by both spouses.

A further distinction in the framing takes into account the common stages of agricultural development – early, middle and late – drawing on insights from Peter Timmer. In the early stages, the key concerns are to ‘get agriculture moving’ (Mosher’s memorable phrase) by putting underused land and labour to work. As the overall economy develops, agricultural development increasingly needs to become efficient and raise productivity to allow labour, and often some capital as well, to be transferred

to other sectors. In late stages, agriculture becomes a minor activity in an economy dominated by urban activities. Processes differ through these stages, as do the public policies and investments needed to support them.

Drivers of agricultural transformation may serve to stimulate or support inclusion, or to prevent and reduce it. Whether they make an important contribution to the inclusiveness of the transformation in a specific case is an empirical question. This study aims to identify clusters of drivers and see how they worked in previous cases of transformation, thus illuminating what might be the next steps in economies still in the early stages of the transformation process.

1.3.2 Lessons from Egypt, Peru and ThailandIn the early 1960s, Egypt and Thailand were largely agrarian economies in which most people lived in rural areas, and engaged either directly in farming or working in agricultural supply chains. Both had low average incomes, with most of the population living in poverty and especially so in rural areas. Population growth was strong: more than 3% a year in Thailand.

Peru at that time was more urbanised with a substantial industrial and services economy and had achieved middle-income status. Peru,

Figure 1 Inclusive agricultural transformation: levels of analysis

Structural transformation

Agricultural and rural transformation

Household transformation

Inclusion that empowerswomen and girls

18

however, had marked geographic and social inequalities: the highland Sierra remained heavily agrarian, with the great majority of the rural population living in poverty.

Since the early 1960s, all three countries have experienced economic growth, urbanisation and industrialisation, agricultural and rural development, and a slow-down in population growth. In the process the material conditions of rural women and girls have improved dramatically.

Thailand: strong economic growth with transformationThailand is the exemplar among the three, despite some mistakes in public policy. The Thai economy has grown rapidly since the early 1960s (Figure 2). In the early 1960s, policy changed from state-led industrialisation to a more liberal approach. In the 1960s and 1970s, industrial growth was based around agricultural processing, but from the 1980s investment by Japanese companies looking for lower labour costs led to the growth export-orientated manufacturing. In the late 1980s a financial crisis pushed the economy into recession, but following bank bailouts and reforms, recovery was quite rapid.

Throughout this period, agricultural grew ahead of population growth, not only feeding a growing population, but also generating net

exports that allowed imports of capital goods. Agricultural development was largely extensive before the 1990s: new land, much of it taken from the forests, was brought into cultivation.

The economy grew from the early 1960s to the late 1980s largely by putting readily available resources to use: the workforce that was expanding through population growth and gaining human capital, and the land that could be converted to fields and pasture.

Economic growth was encouraged by policies and macroeconomic management that for most of the period encouraged private investment, including allowing Thais of Chinese origin to exercise their considerable business skills; public investments in roads, irrigation schemes, health and education; and, during the 1960s and 1970s, aid from the United States to counter insurgency that included funds to construct roads deep into rural areas, providing increased connectivity.

Within 30 years, Thailand’s economic structure was transformed. Agriculture, which in 1961 made up 36% of gross domestic product (GDP), was less than 10% by 1993, a share that it has maintained ever since. Industry’s share of the economy rose from 19% in 1961 to 39% by 1991.

In the rural economy, it was not just agriculture that grew: the rural non-farm economy (RNFE) also expanded, and faster than

Figure 2 Economic growth, Thailand, 1961–2015

Source: Compiled from World Development Indicators

-15%

-10%

-5%

0%

5%

10%

15%

0

1,000

2,000

3,000

4,000

5,000

6,000

1961

1963

1965

1967

1969

1971

1973

1975

1977

1979

1981

1983

1985

1987

1989

1991

1993

1995

1997

1999

2001

2003

2005

2007

2009

2011

2013

2015

Annu

al g

row

th (G

DP p

er c

apita

) GDP per capita (US$ constant, 2010)

19

agriculture. By 2015 the RNFE provided around half of all rural jobs. Some activity is linked to agriculture, such as rice mills, sugar refineries, workshops manufacturing and repairing agricultural machinery, and transport services. Other local businesses have grown to meet the demand from farmers for goods and services.

Economic growth with urbanisation has been accompanied by migration from rural to urban areas to work in assembly plants, construction, domestic service and other jobs such as retailing. Surveys show that among low-income rural households, most have at least one migrant member. Migrants tend to be young, and young rural women migrate at least as commonly as men, if not more so. Most workers in manufacturing plants, and almost all those in domestic service, are women.

Thailand has been through a remarkably rapid demographic transition. Population growth of more 3% a year in the early 1960s had slowed to 1% a year by 1990 (Figure 3). Fertility rates fell rapidly from the early 1970s onwards, as contraception was widely adopted (Figure 4).

The strong uptake of contraception resulted from the family planning programme, run by the Ministry of Health and an NGO. Village women were trained as paramedics to provide advice and contraceptives at the doorstep. Once the programme was established, the adoption of contraception became widespread within a decade, indicating a high underlying demand.

With slowing population growth, out-migration, and more non-farm work, the farm labour force ceased to grow from around 1990 and thereafter declined. Agricultural labour productivity rose in response. It had been stagnant or declining in the 1970s and 1980s, but from the early 1990s it doubled within a decade (Figure 5).

With competition for labour, farm wages tripled in real terms between 1985 and 2003, and have increased subsequently. As wages have risen, poverty has fallen. In 1988 it was estimated that almost 75% of the rural population was living below the national poverty line; by 2013 it was down to less than 14%. The gap between urban and rural rates of poverty has also narrowed, from 30 to just 7 percentage points (Figure 6). Income inequality has also fallen: the Gini

coefficient peaked in the early 1990s at 0.48: by 2013 this had come down to 0.38.

Changing lives of rural women and girls in ThailandThe lives of rural women and girls have been profoundly affected by these economic and demographic changes, as well as by the health and family planning services in rural areas.

Figure 3 Fertility, Thailand, 1961–2014

Source: Compiled from World Development Indicators

1

2

3

4

Fert

ility

rate

5

6

7

1961

1966

1971

1976

1981

1986

1991

1996

2001

2006

2011

Figure 4 Use of contraception, Thailand, 1970–2012

Source: Compiled from World Development Indicators, MICS, 2012

10%

30%

50%

70%

90%

1970

1974

1978

1982

1986

1990

1994

1998

2002

2006

2010

Cont

race

ptiv

e pr

eval

ence

(any

met

hod)

, w

omen

age

s 15

–49

20

Figure 5 Agricultural productivity, Thailand, 1961–2013

Source: Compiled from FAOSTAT data

200

400

600

800

1,000

1,200

1,400

1961

1963

1965

1967

1969

1971

1973

1975

1977

1979

1981

1983

1985

1987

1989

1991

1993

1995

1997

1999

2001

2003

2005

2007

2009

2011

2013

US$

cons

tant

200

4/06

per

hec

atre

, or

agr

icul

tura

l em

ploy

ee

Prod Value/Agri LabValue prod/ha

Figure 6 Poverty, Thailand, 1988–2013

Source: Poverty and Equity database, The World Bank. Poverty headcounts are at nationally-defined poverty lines.

0%

20%

40%

60%

80%

1988

1990

1992

1994

1996

1998

2000

2002

2004

2006

2008

2010

2012

Head

coun

t

Poverty, rural Poverty, urban

Figure 7 Women’s employment, Thailand, 1980–2013

Source: Compiled from World Development Indicators; primarily employment

0

5

Mill

ions

10

15

201319901980

ServicesIndustryAgriculture

31

14

5

2

13

9

4

8

21

As producers, diversification of the rural economy has created new options for women, either as migrants or as workers in rural enterprises. Nationally, a clear picture emerges of women leaving agriculture – much more so than men – and taking up employment in industry and services (Figure 7).

Women’s wages have risen and the gap between men’s and women’s wages has narrowed.

As carers, rural women’s work has been eased by reduced fertility and hence fewer children to care for, and by electricity and running water in homes. Despite these improvements, in rural Thailand women still appear to do far more domestic work than men. A national survey in 2001 reported that in rural areas, women spend an average 5.4 hours a day on domestic tasks and child care, compared to 2.9 hours for men. Moreover, as parents age, daughters are expected to take care of them. Young migrant women are expected to send remittances much more than migrant men are.

As people, public investments in health and education mean that rural women and girls are healthier and better educated than their mothers and grandmothers were. Female life expectancy has risen from 57 years in 1961 to 78 years in 2014. Under-five mortality has fallen from 147 per 1,000 in 1960 to just 13 in 2014.

Thailand already had a high rate of primary school enrolment in the 1960s: it was almost universal by 1990. Before 1990, however, secondary schooling lagged: only 30% of eligible children were enrolled. The belated recognition of this deficit led to heavy public investment from then on, with up to 25% of the public budget spent on education (Fan et al., 2004). Consequently, by 2013, 85% of boys and 89% of girls were in secondary school. By 2012, there was almost no difference in enrolment between urban and rural areas.

As agents, women have been restricted by gender norms that ascribe women a secondary role in a society where men are expected to lead and take key decisions. Three norms, however, have allowed women to take up opportunities.

One, most groups in rural Thailand have matrilineal inheritance and matrilocal residence upon marriage. This may not entitle women to

land and other property, but it nuances any sense that only men own property.

Two, Buddhist ideals that stress the non-material, have meant men have preferred to leave business and household finances to women. Hence, when there have been opportunities to participate in rural non-farm enterprises, women have been reasonably well placed to take them.

Three, most important of all, norms have not prevented young rural women from migrating: even if their unsupervised travel is not well regarded, it has been tolerated.

Thailand’s modern history shows how much progress can be made when broad-based growth takes place, with spending on rural public services, in a context where gender norms, while favouring men, do not block women’s progress. It also shows how beneficial a rapid demographic transition can be for rural transformation and women’s lives.

Peru and Egypt: similarities and differences

PeruPeru dissipated its earlier development through losing almost 25 years of economic growth between the late 1960s and early 1990s. Unsuccessful land reform, adverse movements of commodity prices, unwise macroeconomic policy and civil war in parts of the Andean highlands meant that by 1992 the average income per person was no higher than it had been in the early 1960s (Figure 8).

Since the early 1990s, Peru’s economy has grown rapidly, initially on a narrow base of mining, and exports from commercial farming along the coast. In the 1990s that meant too few jobs were created, and poverty and inequality rose. But in the 2000s a combination of slowing population growth and job creation – including in export agriculture, the value of which increased more than 11 times in real terms from the mid-1990s to 2012, and created another 60,000 jobs in the 2000s, a large number for an agricultural workforce of 2.5–4 million – has raised employment and wages. This, plus public spending on roads, irrigation, health, education and cash transfers in low-income areas (predominantly the highlands) has led to major

22

reductions in poverty and inequality since the early 2000s (Figure 9).

The material conditions of life for rural women have improved substantially. Indeed, recent life histories indicate a sea-change in the perspectives of rural parents for their daughters. Seeing new off-farm opportunities,

they recognise that that secondary schooling is critical for girls’ prospects.

As in Thailand, although the material conditions of rural women’s lives have improved, changes to gender norms and status have been slower.

Peru confirms that a combination of a tightening labour market – caused by declining

Figure 8 Economic growth, Peru 1960–2015

-20%

-15%

-10%

-5%

0%

5%

10%

15%

Annu

al g

row

th

2014

2012

2010

2008

2006

2004

2002

2000

1998

1996

1994

1992

1990

1988

1986

1984

1982

1980

1978

1976

1974

1972

1970

1968

1966

1964

1962

1960

2.5

3.0

3.5

4.0

4.5

5.0

5.5

6.0

GDP per capita (US$ thousands constant, 2010)

Figure 9 Poverty and inequality, Peru, 1985–2015

Source: Poverty and Equity Database, World Bank. Poverty measured by national poverty lines.

20

35

50

65

80

1985 1987 1989 1991 1993 1995 1997 1999 2001 2003 2005 2007 2009 2011 2013

Perc

enta

ge p

oor/

Gini

sco

re (%

)

Gini coeff. Rural poverty National poverty

23

labour force growth and increased labour demand – and investment in people, including reproductive health and family planning programmes, can do much for rural women and girls. It also shows how a country can recover from economic recession and civil war.

EgyptEgypt provides a further contrast. Economic growth, in which agriculture has played a significant role, has fluctuated since the 1960s, resulting in a worse performance than Thailand. Nevertheless, GDP per person in 2015 was 3.5 times the level estimated in 1965 (Figure 10).

The differences to Peru and Thailand, however, stem from serious challenges to inclusive growth that reduces poverty, with consequences for rural women who have made less progress than in the other two countries. Three interrelated factors account for much the difference.

One, Egypt’s growth model has not created enough jobs to meet demand, above all since the mid-1990s. It is not as though the government has not tried in the last 60 or so years: state-led industrialisation; redistribution of land to tenant and landless farmers in the 1950s; and a succession of huge public construction projects, starting with the Aswan High Dam in the 1950s and continuing with desert-irrigation schemes, all created significant employment. But in recent decades, the employment elasticity of growth has fallen. Despite the economic liberalisation that might have been expected to attract foreign investors, Egypt has not seen the growth of assembly plants and light engineering in the ways that so many Asian emerging economies have. Moreover, many of the jobs that have been created in Egypt have been regarded as male jobs, providing few opportunities for women. For instance, tourism, in many countries a major employer of women, generates hardly any jobs for women in Egypt. This reflects the interplay in Egypt of the restrictive gender norms regarding economic activity, and the lack of labour demand. While not sufficient, the examples of Thailand and Peru do show that high growth in demand for labour in nonfarm sectors helps shift economic gender norms toward more opportunities for women.

While agricultural development has been largely successful in raising output per person and incomes in rural areas, given the very high ratio of population to arable land, the bulk of additional jobs in rural Egypt have to come from off the farm.

Two, Egypt is going through a prolonged, and possibly even stalled, demographic transition. Fertility rates, even in urban areas, have failed to fall toward natural replacement levels and remain stubbornly high (Figure 11). While this continues, Egypt faces the prospect of labour force expanding by 2.9% a year: a rate that has easily outpaced the growth of jobs in most years since the mid-1990s. Not only is this condemning youth to a difficult search for employment that is often informal and ill paid; but also, it disproportionately affects women, and especially rural women, who are less educated and less mobile than their urban counterparts.

Three, the lack of acceptance of women’s freedom of movement severely curtails their job prospects, above all, for rural women. Urban women may be able to find a job close to their home where it is acceptable for them to make a short journey; that is far less likely in a village.

Rural girls thus face the prospect that their future lives will be restricted to the domestic domain where they will be expected to do much of the farm work, bear and care for children, and be responsible for domestic chores. And to do so with few firm rights to property and under the threat of violence from husbands.

All that said, poverty and deprivation are the still major determinants of life opportunities and options, more so than gender. To be born into a poor rural household in Upper Egypt, perhaps landless, to parents of little education and few formal skills, makes the odds of escaping poverty low. This applies even more so to girls, but the fate of the boys is not so enviable either. In such conditions, the search for gender equity goes hand in hand with poverty reduction.

In sum, by and large, rural women’s lives have improved in all three countries through structural and rural transformation, although notably less in Egypt. Rural girls face better economic prospects than their grandmothers did back in the 1960s and 1970s. Progress, however, is uneven.

24

Rural women have better economic opportunities, and are likely to be better rewarded for their efforts, especially in countries where non-farm employment growth has been significantly higher than labour force growth. Low non-farm employment growth seems to reinforce restrictive segregation of

economic activity. Rural women are generally healthier – and less likely to have suffered from incapacitating malnutrition in infancy – and most have control over their fertility. Most have been to school, and increasingly they will have been to secondary school. These are the clearest signs of progress seen.

Less clear are changes to norms by which the bulk of household chores fall to rural women, as well as almost all child care. Fewer children, piped water on premises and rural electricity may well have reduced some of the chores, but the gender disparity remains wide.

Progress on equity whereby rural women can exercise their full rights as citizens, the equal of men, is slow. Women’s status may be rising, but in rural areas they usually come second to men.

1.4 Ethiopia and Tanzania

These two countries are similar in being low-income, still predominantly agrarian. Both share a checkered economic past as their economies faltered in the 1970s and 1980s owing to failed state-led economic development. Since then, however, both countries have seen economic growth (Figures 12 and 13) with an increasing share of output from non-farm activities, albeit mainly services.

Figure 10 Economic growth, Egypt, 1967–2015

Source: Compiled from World Development Indicators

-10%

-5%

0%

5%

10%

15%

Grow

th p

er c

apita

(per

yea

r)

2015

2013

2011

2009

2007

2005

2003

2001

1999

1997

1995

1993

1991

1989

1987

1985

1983

1981

1979

1977

1975

1973

1971

1969

1967

500

1,000

1,500

2,000

2,500

3,000

GDP per capita (US$ constant, 2010)

Figure 11 Fertility, Egypt, 1988–2014

Source: Compiled from DHS

2.0

3.0

4.0

Fert

ility

rate

5.0

6.0

1988

1990

1992

1994

1996

1998

2000

2002

2004

2006

2008

2010

2012

2014

Rural Urban

25

Both Ethiopia and Tanzania have improved macroeconomic policy and therefore stability; the investment climate has been enhanced through infrastructure development and more business-friendly regulations and taxes; the role of the state in production has been curtailed; and both countries have become more integrated into the global economy through lower tariffs and non-tariff barriers.

In Ethiopia and Tanzania, clearly the lives of rural women and girls have improved since the 1990s. Economic growth and structural transformation, with increases in agricultural productivity – no matter how modest and patchy – mean that returns to rural labour have risen. Moreover, the growth of the rural non-farm economy (RNFE) and improved connections from rural areas to secondary and primary cities has widened the range of opportunities for rural labour.

Public provision of education, health care and clean water means that rural people in general, and women and girls in particular, have not only have seen significant improvements in their welfare; but also, that they are better able to take up the jobs created by growth.

Fertility has fallen in both countries, even if the reduction in Tanzania has been limited to date, which has reduced the amount of child care that rural women have to provide. In some rural areas, improved water supplies have cut the time taken to draw water, a task that generally falls to rural women and girls.

Norms that see a woman’s place as primarily a wife and mother, that assign rights to property first and foremost to men, and that tolerate worse treatment of women than men – including domestic violence perpetrated by men – does not seem to have changed as much.

Progress for rural women, then, has been variable. Depending on which of the different dimensions of women’s empowerment are examined, it is possible to see promising progress; or to see stasis punctuated by occasional, but limited, improvements in material conditions. It is hard to be more certain of the dynamics of changes to the lives of rural women and girls, partly owing to the limitations of the statistics available, and partly to the difficulty of predicting future trajectories when the key statistics are only now beginning to show changes.

The longer-term studies of Egypt, Peru and Thailand suggest that it is only with the benefit of hindsight that remarkable progress can be seen. Some observers of rural Thailand in the 1980s, for example, doubted that sustained development that would favour ordinary rural people was underway. The sceptics focused on what had previously been important in the villages – access to land, rice production, for example – and failed to realise the importance of the growth of the RNFE, the proliferation of urban opportunities, girls going to school and falling rural fertility.

Rather than attempt in vain to produce a reliable scorecard of progress toward gender equity in rural Ethiopia and Tanzania, it may be more informative to pick out the key areas that will probably determine how much progress is seen in the future. Three processes above all will determine progress. One is the rate of growth of the urban economy, manufacturing and the rural non-farm economy. Most of the above-average productivity, and hence usually higher wages, will be in these sectors. Agricultural development still matters since so many people depend on farm incomes and will continue to do so in the medium-term.

A second concerns family planning, uptake of contraception and rural fertility. Not only can family planning empower rural women, but the sooner that fertility falls toward replacement levels, the sooner the growth in labour supply declines, the sooner demand for workers will outstrip growth of the labour force so that wages and earnings start to rise significantly. Although poorer and with lower female education levels, Ethiopia has made better progress owing to an effective government strategy, with leadership from the highest levels.

The third critical process is girl’s secondary schooling, which supports women’s empowerment, not only by giving young women the qualifications that most formal employers seek and expect; but also by making early marriage less likely, and by raising young women’s status.

1.5 Conclusions and discussion

Four main findings emerge from this study.

26

Figure 12 Economic growth, Ethiopia, 1981–2015

Source: Compiled from World Development Indicators

-20

-15

-10

-5

0

%

5

10

15

20

GDP growth

2015

2013

2011

2009

2007

2005

2003

2001

1999

1997

1995

1993

1991

1989

1987

1985

1983

1981

100

150

200

250

300

US$ (constant, 2010)

350

400

450

500

GDP per capita

Figure 13 Economic growth, Tanzania, 1988–2015

Source: Compiled from World Development Indicators

%

GDP annual change

US$ (constant, 2010)

GDP per capita

0

2

4

6

8

10

2015

2014

2013

2012

2011

2010

2009

2008

2007

2006

2005

2004

2003

2002

2001

2000

1999

1998

1997

1996

1995

1994

1993

1992

1991

1990

1989

1988

400

500

600

700

800

900

27

One, broad-based development can result not only create many additional jobs to meet the rapid expansion in the labour force that arises as countries undergo a demographic transition; but also, jobs where average labour productivity rises as well.

Agricultural development is part and parcel of such growth and transformation. In the early stages of development in the 1960s and 1970s, presented in the three long-term case studies, agriculture largely met the Johnston-Mellor functions of feeding the nation, exporting to earn foreign exchange, raising rural incomes to provide domestic demand for the output of urban and rural non-farm industry and services, and releasing labour and capital to the growing urban economy.

It is not just agricultural development that counts in rural areas, however: the growth of the RNFE also matters. Since the RNFE is not a category in national accounts and is infrequently, unevenly, and idiosyncratically surveyed, reliable information on how it grows is sparse. Surveys of rural households, nonetheless, report large and increasing shares, often more than half of rural incomes, coming from sources other than farming. Non-farm activity grows through two mechanisms: activities that are linked to agriculture either directly in the supply chains upstream and downstream of the farm, or indirectly through spending by farm households; and through public spending on rural infrastructure and services that creates rural jobs.

Even though smallholder farming and the RNFE tend to be quite labour-intensive, migration from rural to urban areas can play an important role in raising earnings. Not only does this provide labour for an expanding urban economy, but it also raises returns to household labour.

The generation of rural jobs, and migration options, seems especially important for rural women. Within agriculture, women tend to farm crops and livestock at low labour productivity. Often, they are also deprived of access to inputs and technical innovations. Compared to such work, a job in the RNFE or migrating to the city can pay much more. Moreover, while working within the farm household, a woman may see little cash income or her husband may appropriate it. When working for wages or

operating a small enterprise, a woman is much more likely to control her income. And with that control can often come greater status and self-esteem.

A simple proposition emerges from these cases: getting out of agriculture matters more for rural women than for rural men.

Two, rural people can take up the better opportunities economic growth creates only if they are healthy, literate and numerate. Hence public investments in rural education, health, and clean water and sanitation make a real difference. Primary health services, water and sanitation, have done much to reduce disease: the decline in under-five mortality in all five countries, taking this a proxy for overall rural health, has been remarkable. In a generation, the pain of losing an infant has gone from being commonplace to a sad rarity.

Schooling in rural areas has increased notably in all five countries: for the middle-income countries, high rates of enrolment in secondary school are common, with little difference between girls and boys.

Three, one service matters more than expected before reviewing these cases: family planning. When family planning is readily accessible to rural women – which usually means services in the village, preferably on the doorstep, and provided by people they trust, in an overall supportive cultural environment, contraception is usually rapidly and widely adopted. Not only does this put women in control of their fertility, but also it leads to fewer births, and – with a delay of a decade or two – a pronounced slow-down in the growth of the labour force. Once rural populations cease to grow rapidly, labour shortages arise, investment to raise labour productivity is encouraged, and wages for unskilled rural work rise.

Four, gender norms count. They may not be set in stone, often being reinterpreted when the economic advantage is obvious – as in rural Thailand when it became clear in the 1970s that young rural women could earn valuable incomes if they migrated – yet nevertheless norms can either facilitate or block these processes. Mobility stands out: rural women have far more options when it is acceptable for them to travel independently, to visit market centers, to commute for jobs, or

28

to migrate to the city. Which is not to say that in societies where women’s rights are not firmly established in practice, women who travel do not incur risks in personal safety and poor treatment at work: they do. The point, however, is to reduce those risks, rather than prevent movement.

Finally, even if some of the strongest drivers of change for rural women and girls are not specifically agricultural, this does not mean that the considerable interest in women in

agriculture awakened in the last ten or so years has been in vain. Far from it. Women farmers are disadvantaged and much can be done to remedy this.

Rural women’s rights – to land and property, to public services, to fair and decent treatment at work, to protection from violence – need strengthening. All these rights support women in rural areas, and when they seek to leave rural areas, as migrants.

29

2 Introduction

2.1 Motivation: improving the lives of rural women and girls

Many and probably most rural women and girls are disadvantaged in developing countries. Compared to men and boys they receive less formal education, they often receive less health care, they have fewer opportunities to work outside the household and when they do, they are often paid less and treated worse than men. Most rural women are constrained by social norms that define them primarily as mothers and houseworkers. Women are typically expected to be subservient to men. At worst they are subject to male abuse and physical violence.

Not only is this unjust, but also it means the full potential of rural women and girls – as people, workers, citizens, leaders – is not realised, to the detriment of their households and families, their rural communities and indeed their nations as a whole. When rural mothers lack the basics of life – food, income, health care – their children are at risk. Gender inequality thus threatens future generations.

While the current reality is dismal, great opportunity lies in correcting unfairness and allowing rural women and girls to realise their potential. Hence the question of how to improve the lives of rural women and girls is central to rural and national development.

While the answers may be neither simple nor straightforward, change is possible, and indeed has taken place. The lives of rural women and girls are changing. In many developing countries some aspects of the lives of girls and young women are far removed from those of their grandmothers. It is much more common today for girls in rural areas to go to school, for pregnant women to receive ante-natal care, for women to take on paid jobs and run businesses, and for rural women to migrate to towns and cities for better opportunities. Lessons may be

learned from the many and diverse experience of such changes worldwide.

These changes arise within the overall development of developing countries, some of which have seen profound transformations in their economies and societies over the last 50 years. Many formerly low-income countries have gone from being largely agrarian and rural to industrialised and urban, and in the process have seen incomes rise and living standards improve. They have also undergone a demographic transition, from high to low rates of birth and death, during which their populations have grown rapidly.

Rural areas and their dominant activity of agriculture may be relatively less important than they were, but it is these areas that experience some of the largest development deficits: poverty, hunger, low education, and poor health. Against this backdrop, this report looks at how the lives of rural women and girls can be improved.

2.2 Aims of study, questions addressed

Two questions were set for this study at the start by the Bill and Melinda Gates Foundation:

• What does an inclusive agricultural transformation that empowers women look like?

• What are the implications for: (a) policy; (b) infrastructure; (c) technology; and (d) behavior change?

The first question refers to agriculture, rather than rural development overall, because agriculture tends in most rural areas of low-income countries to dominate the rural economy – even if most rural households typically derive one third or more of their incomes from non-farm activity (Haggblade et al., 2007).

30

It reflects the considerable interest that has been shown in the gendered aspects of agriculture in the new century (FAO et al., 2010; FAO, 2011; World Bank and ONE, 2014). Rural women’s disadvantage manifests itself in agriculture, as women farmers struggle with less secure access to land, poorer supplies of inputs, less access to credit and technical information than men farmers.

We address this question, however, not at the detailed level of how to overcome women farmers’ disadvantages – to which a burgeoning literature exists; but at a higher level that considers the consequences of agricultural transformation – in which higher labour productivity is central – for rural women.

2.3 Approach and methods

The approach beings by setting out a framework to consider how development and transformation, seen at the national level, for agriculture and the rural economy, for rural households and for women, may affect the lives of rural women and girls.

Then, in the light of what the framework revealed, changes for rural women since the 1960s in Egypt, Peru and Thailand were studied. All three countries have seen economic growth, urbanisation, and a marked shift in their economic structure as agriculture has declined relative to industry and services. Available statistics were collected and existing literature

was reviewed to provide the material for interpretive studies that set out both the main changes in national and rural economies, and what has changed for rural women and girls.

Finally, two contemporary cases were examined: Ethiopia and Tanzania, both low-income countries that are still largely rural and agrarian, but with growing economies and transformations in progress. For Ethiopia, the study relied mainly on existing statistics and literature, complemented by an exploration of policy for women’s empowerment. Field work had been contemplated, but could not be carried out owing to insecurity in rural areas during 2016.

For Tanzania, the analysis of statistics included new tabulations of data from the 2014 Labour Force Survey in order to get a understanding of employment and livelihood changes. In addition, a qualitative study of change for rural women and girls was undertaken in two selected, contrasting areas of the country: Kilimanjaro Region and Morogoro Region. This was complemented by insights from other ongoing qualitative field work being undertaken by a team led by Dan Brockington (2018).

The rest of this report follows the three elements outlined: chapter 3 sets out the framework; chapter 4 reports findings from the three cases of long-term change; chapter 5 presents the insights from Ethiopia and Tanzania; and chapter 6 concludes by highlighting the main findings and discussing their implications.

31

3 Agricultural development and the lives of rural women and girls: a framework

1 See Booth, McMillan, Page and te Velde (2017) for a discussion of this definition and its importance to growth and development.

2 Appropriate segmentation varies by country, but common distinctions comprise individual characteristics and identity (ethnicity, gender, religion, etc.); location (rural, urban, particular regions, etc.); and economic class.

3 See Alkire et al. (2013) and Fox and Romero (2016) for a discussion of definitions of economic empowerment.

3.1 Introduction to the framework

Economic transformation – the movement of factors of production to higher-productivity activities, both within and between sectors1 – not only goes hand-in-hand with growth, but also can help make growth broad-based, inclusive, and more resilient to shocks.

Economic transformation encompasses both horizontal structural change as resources move between sectors, and vertical structural change as labour productivity rises within a sector. Both processes usually involve investment of more capital per unit of labour, adoption of better technology, and more efficient allocation of resources. Trade can help foster these changes. It allows specialisation of production, using exports as the means to overcome the limited size of domestic markets. Imports of capital goods and foreign direct investment (FDI) can be ways to obtain improved technology and know-how.

Ideally, economic transformation is socially inclusive, creating new choices and opportunities for all. In reality, some sectors and economic groups may move ahead faster.

The policy challenge is to ensure that growth and transformation are as inclusive as possible, that no segment of society2 is made worse off. Although the measurement of the benefits usually involves money metrics such a household consumption levels, increasingly the concept of inclusive growth has come to encompass improvements in (or at least no loss of) human rights and individual’s ability to independently exert control in their economic life and make choices according to what they value (economic empowerment). This is especially the case where these have been denied or restricted by the institutions of society – the legal system, the social leaders such as local elders, or cultural traditions and social norms, especially those rooted in identity. Inclusion is by definition a medium-term concept, as there may be short-run setbacks caused by economic and other types of shock which temporarily halt the process.3

The agriculture sector, and rural areas in general, are key elements in transformation. In low-income countries, agriculture dominates rural economic life. Rural areas are where the poorest and most excluded people tend to live.

32

Agriculture is typically the lowest productivity sector in developing countries, characterised by low earning and vulnerable employment.

Agricultural transformation arises when labour productivity on farms rises significantly. This then allows some farm labour to move to more productive activities, either locally in the RNFE, or in towns and cities. It is usually associated with commercialisation of agriculture, as an increasing share of farm output is sold and as more inputs are brought in: more efficient rural markets are therefore also part of agricultural transformation.

Rapid agricultural transformation can speed up the overall pace of economic growth and transformation, by providing food to growing urban areas, and earnings from exports to allow the import of capital goods for other sectors. Multipliers operate in rural areas as well. Additional jobs are created in agricultural supply chains both in provision of inputs and services, as well as in processing, transport and storage of produce. As farmers spend increased incomes, even more jobs may be created providing the goods and services they demand. These effects provide rural households with more opportunities to diversify their income and to earn more. No country has significantly reduced rural poverty without agricultural transformation.

Households – where many of the basic economic decisions of individual’s lives such as when to work, where to work, what to spend, how much to save, what to buy and whether to sell or otherwise dispose of assets or income – are the basic economic unit of inclusiveness. The household can also be a scene of exclusion, especially for women, since many gender norms apply first and foremost within the household. Household welfare, measured by income, consumption and asset levels, provide the basic metrics of inclusiveness.

In rural areas, this would cover income from farm and non-farm economic activities as well as in-kind income from publicly provided or subsidised social and infrastructure services (e.g. education, health care, water supply, roads and transport). Assets such as land, human capital, housing and consumer durables, livestock, and savings, both support households in their income-earning and unremunerated activities and

make consumption-smoothing possible in cases of shock.

Inter-household inequalities in economic opportunities, earnings, or monetary or non-monetary consumption are commonly found, arising from differential endowments, talent, preference for work and good fortune. When these inequalities become large, however, not only may some disadvantaged households not share in the benefits arising from growth and transformation, but also this may hinder future growth (World Bank, 2006).

Processes of rural transformation do not necessarily benefit everyone. Households may lose out when economic opportunities and provision of public services are geographically concentrated, so that some areas do not benefit. They may lose out when services are provided to some social groups, but not others. Even more serious is when some households are dispossessed of their access to land and water by rapacious investors, state agencies or other farmers who have the power to usurp their rights.

Even if the monetary welfare of the household improves, it is not guaranteed that all household members have the same standard of living or that everyone’s welfare improves. Households are complex units, and resources are seldom shared equally. Power dynamics within the household may constrain the opportunities and choices of individual members. For example, if women bear the responsibility for most of the domestic chores, and no investments are made in labour-saving assets (such as running water, more efficient kitchen stoves or a refrigerator), women will be unable to take advantage of the new income earning opportunities that rural transformation brings, or they may lose leisure time if they do. Less powerful individuals may hesitate to take up new opportunities if these challenge current power dynamics, and therefore could trigger a backlash, making them worse off. These individuals are not empowered to exert control over or influence economic or other decisions in the household which affect their welfare, or that of their dependants.

Intra-household economic inequalities arise primarily because of dependence, although they may be reinforced by the prevailing institutions (laws, norms, and traditions that govern

33

behavior). Development economists have tended to focus on gender inequalities, and principally on women and girls in low-income countries. Inequalities are, however, determined by other factors, such as age (being younger or older than the most powerful person in the household), or because of a more distant relationship to this person (e.g. a stepchild, or an orphaned niece or nephew). Intra-household inequalities may create inter-household inequalities, and they may hold back transformation if not all of its members are pursuing opportunities from which they would all benefit.

Gender inequality is central to this study, hence a gender inclusion is a particular interest. As will be set out in more detail in section 3.5, agricultural and rural development will be gender-inclusive if increases in household income raise the consumption, capacity and agency to act of all members, above all women and girls.

While inclusive economic, agricultural, and rural transformations are therefore desirable and often sought by development actors, they are not always achieved. Measured at the household level, rural poverty has been stubbornly high in many African countries, for example, despite evidence of economy-wide structural transformation, in part because productivity improvement in agriculture has been too slow.4 Other countries and regions have experienced inclusive agricultural transformation, but the lives of rural women and girls have hardly changed, as measured by health and nutrition indicators or movement into higher-value economic activities. As IFAD (2016) notes, ‘inclusive rural transformation must be made to happen; it will not happen automatically’ (IFAD, 2017: 16). But what does that mean? How does inclusive agricultural transformation happen? Is it a question of initial conditions or government policies and programmes? How do these two interact?

Given the myriad factors involved in determining the outcomes of structural and rural transformation processes and their interdependence, it is not possible to develop a universal causal model to analyse all low-income

4 See IFAD (2016) for an empirical analysis of these cases

economies. The approach used here is one of case studies of transformations. To use this economic history to shed light on these questions, we developed a framework to organise the historical evidence.

Our framework identifies the processes which need to work on multiple levels of the economy to achieve the sustainable improvements in economic outcomes desired:

• the macroeconomic or national level (national income, productivity and productivity gaps, structure of employment)

• the sectoral level (agricultural productivity; commercialisation)

• the rural household and community level (household income, asset, and welfare levels; empowerment within the community and society)

• the individual level (individual income, asset, and welfare level; human rights and individual empowerment).

Our hypothesis is that within the transformation processes at each level, there are drivers of improved outcomes. These drivers may be initial conditions or events external to the level of analysis (such as favourable terms of trade or global/regional political environment) at the national level; good and generally predictable climate; climate poorly suited for plantation agriculture, helping ensure an equitable distribution of land at the household level; improvements in health technology allowing better access to health care for all at the individual level. Or they may be policies and/or programmes, such as a cost-effective rural roads programme or a land-titling programme which ensures that titles are jointly held by both spouses. They may also be norms that can prevent women from benefiting from changes (Bishop, 2017).

These drivers may either stimulate or support inclusion, or prevent or reduce it. Whether they are important contributors to the inclusiveness of the transformation in a specific case is an empirical question. Our goal is to identify

34

clusters of drivers and see how they worked in previous cases of transformation, thus illuminating what might be the next steps in economies that are still in the early stages of the transformation process.

Our historical method, which we apply to case studies of countries already transformed, is therefore a set of nested questions on the country experience at each level (see Figure 2.1). These questions come mainly from the existing literature on structural and agricultural transformation, pro-poor and inclusive growth, and gender equity and women’s economic empowerment. We examine the data at each level in each country using a set of standard indicators to (a) quantify the extent of the transformation and its inclusiveness, and (b) look for the presence or absence of the drivers and processes we have identified.

The rest of this chapter introduces the indicators we use to measure the transformations at the four levels, lists important drivers of change found in the literature review, and suggests some ways in which these drivers interact with initial conditions in the country to create a more or less inclusive process of transformation.

3.2 National structural transformation

It is widely agreed that transformation from a fundamentally agrarian, subsistence economy to one that is urban, integrated, and enterprise-dominated is the essence of economic development, and is what sustains improvements in economic welfare. This has been the experience throughout the industrialised world as documented by Duarte and Restuccia (Duarte and Restuccia, 2010). Moreover, since the work of Arthur Lewis in the 1950s structural transformation is generally considered a key element of economic development. Since his initial analysis (Lewis, 1954), the progress of this transition has been measured in output, productivity and employment. Specifically, following McMillan et al. (2014) and Timmer et al. (2012), we measure the extent of structural transformation by:

• the share of output and employment in the higher productivity (non-farm) sectors, and

• growth of labour productivity in the economy, especially in the non-farm sectors.

Drivers of changeThere is a long literature on what drives structural transformation. Most analysts have focused on the role of industrialisation (Duarte and Restuccia, 2010; Rodrik, 2015; Herrendorf et al., 2013), owing to the spillovers to other sectors of the economy from development of this sector. But this begs the question – what spurs industrialisation? Key factors identified based on case studies of OECD and other East Asian economies tend to focus on policies which raise savings and reduce investor costs and/or risk so that new investment to raise labour productivity continuously takes place. These drivers include:

• macroeconomic and political stability (to reduce investor risk)

• increased integration into the global economy in order to bring in technology and create a larger market for production – needed because manufacturing plants are often subject to economies of scale

Figure 14 Achieving Inclusive Rural Transformation: Levels of Analysis

Structural transformation

Agricultural and rural transformation

Household transformation

Inclusion that empowerswomen and girls

35

• an adequate business environment, including the legal institutions needed to reduce transaction costs in order to increase the returns on investment; (contract enforcement, protection of private property and the returns on investment, etc.)5

• financial deepening and inclusion, to facilitate savings and reduce the cost of capital for new investment

• adequate supply of entrepreneurial aspirations and skills

• public investment in complementary social and infrastructure services with important externalities such as human capital, transport, water and sanitation, energy, and public safety

• ‘competitive cities’, which create agglomeration economies, support rural migrants to establish themselves, and facilitate job-seeking the livelihoods of a growing urban population

• a speedy demographic transition, which creates a ‘demographic dividend’ through a declining dependency ratio, which tends to increase public and private savings.

Structural transformation means more production in enterprises, and less production in households. The process means that the growing enterprise sector will demand more labour. This labour will be supplied by (a) those about to leave school and enter the labour force and those in the labour force but unemployed; (b) those already in the labour force, engaged in home production (farmers who leave their land or cease other forms of self-employment); and (c) those not in the labour force who might be induced to enter it because of new earning opportunities. Movement of labour into enterprises, where wages are earned, is known as the employment transformation.

Employment transformation is the critical link between economic growth, transformation, and poverty reduction. If that movement comes from all of the sources listed above, the transformation will be faster. Lower fertility leading to lower growth of the labour force supports a faster employment transformation by lowering the

5 The business environment need not be great at the start, but tends to improve as the transformation proceeds.

number of people who need to be absorbed into the higher-productivity sectors during the transformation process. Fox and Thomas (2016) identified a lagging employment transformation as a symptom of a less inclusive transformation in Sub-Saharan Africa (SSA). They attributed this in part to the lack of a manufacturing export sector in the output transformation, as well as high labour force growth in slowing the employment transformation leading too much labour to be absorbed in non-farm enterprises.

Enterprises benefit from economic density, so economic transformation and urbanisation usually occur together, implying spatial as well as well as sectoral labour mobility.

Structural transformation is long process, characterised by leading and lagging sectors. The outcome, however, is a ‘post-industrial’ economy, characterised by most of the labour force being engaged in wage labour, covered by a contract and paying taxes (formality), a high level of urbanisation, low birth and death rates, and convergence in sectoral productivity levels and growth rates.

3.2.1 Making structural transformation more inclusiveMajor economic change clearly brings winners and losers. Initial conditions leading to heterogeneous outcomes and unexpected consequences are common. Many of the processes associated with structural transformation may affect the distribution of welfare and opportunities across society. The goal is that they lead to higher welfare for all segments of society, even if some gain more than others. Analysis of historical evidence argues that at least in the initial stages, structural transformation inevitably increases the rural–urban gap in earnings and consumption, driven in part by a widening gap in absolute levels of labour productivity among sectors (Timmer and Akkus, 2008). But this does not mean that the process cannot be inclusive. In East Asia, poverty fell rapidly during structural transformation both among rural and urban households even as inequality widened (Timmer and Akkus, 2008).

36

Economy-wide processes which can help make the transformation inclusive include:

• growth of firms producing labour-intensive goods and services, raising the demand for wage labour, especially formal or year-round wage labour which provides more income security

• slower labour force growth (caused by rapidly declining fertility), which speeds up the employment transformation, which will help raise real wages and decrease the share of the labour force in vulnerable, informal employment

• collective-action processes to support workers’ rights, including gradual expansion of employment-protection legislation

• safety-net policies to support workers’ incomes through the transition and provide a consumption floor for those who may be left behind.

Conversely, less inclusive transformations have been characterised by at least one of the following processes:

• ownership concentration in the enterprise sector, sometimes combined with authoritarian governments, which leads to slow job growth, weak collective-action processes, poor working conditions, and lower pay

• poor urban governance during the process of urbanisation, leading to high urban inequality and slums

• discrimination in employment and pay increases earnings gaps

• lack of investment in connectivity, leading to a concentration of activity in capital cities

• high labour force growth, caused by slowly declining fertility – where cultural norms may prevent faster declines.

Transformations which started as less inclusive – for example in Korea, under authoritarian governments, where ownership concentration and suppression of unions led to poor working conditions in the manufacturing sector in the initial stages of industrialisation – may evolve into more inclusive ones. A completed transformation does not guarantee an inclusive

economy and society. Even in democratic, transformed economies, occupational segregation by gender and gender pay gaps persist, as do earnings gaps by ethnic groups (Dolado et al., 2002). These inclusiveness gaps have not, however, prevented less privileged groups from gaining absolutely in economic terms as the transformation progressed, and in many cases relatively as well.

There are a few countries that have made progress in transforming their economies without improving the welfare of broad segments of the rural population. Likewise, neglect of the agriculture sector in transformation policy leads to a less inclusive transformation (IFAD, 2016). Thus we now turn to the issue of inclusive agricultural and rural transformation.

3.3 Agricultural and rural transformation

Although agriculture’s share of output and employment declines relative to other sectors during structural transformation, agriculture itself has both to increase output – faster than population growth – and become more productive to contribute to that transformation (Breisinger et al., 2011; Duarte and Restuccia, 2010; Herrendorf et al., 2013; Szirmai, 2012; Timmer, 2009).

Transformation sees agriculture move from producing mainly for the subsistence of farming households, with low returns to labour, to a commercial activity, in which much of the output is sold, from farms with higher labour productivity. This allows agriculture to fulfil its prime functions of providing food and agricultural raw materials to the national economy, preferably at relatively low cost; while transferring labour and capital for investment in manufacturing and services. In low-income countries lacking minerals, oil and gas, it is also likely that agriculture will be the mainstay of exports, a key source of foreign exchange for the economy (Johnston and Mellor, 1961).

Key indicators of agricultural transformation are:

• rising share of farm output sold • rising labour productivity in agriculture (value of output per person employed

37

• falling share of the labour force primarily employed in agriculture or livestock production.

The early stages (see Box 1) of agricultural development are demanding: output has to rise and a surplus has to be generated, while the sector loses some of its labour, savings, and human capital to other sectors. Moreover, as output rises, the terms of trade may well move against agriculture since relatively inelastic demand for its output means agricultural prices will tend to fall with increased output, especially for bulky staples where transport costs tend to insulate the domestic from international markets.

This circle can only be squared, without reducing rural populations to penury, if ratios of outputs to inputs rise: that is, if agriculture becomes more productive. In the early stages of development, agriculture often enjoys two advantages that make this possible.

One is the availability of improved technology, as typically manifested in: improved varieties of seed and animal breeds; manufactured fertiliser and agro-chemicals; industrially manufactured tools, including irrigation equipment and farm machinery. Late developing countries benefit from more than half a century of formal agricultural research into improved methods for tropical agriculture, as well as from the existence of industries to supply inputs. Technology from other countries or regions may need some adaptation before it becomes effective in farming systems, but this is significantly simpler than developing the original technology (World Bank, 2007).

The other advantage is that often low-income countries have under-used factors of production that can be put to work in agriculture. Some have land that can readily be converted to fields and pastures. Labour, too, may be relatively abundant owing to seasonal underemployment and to rapid population growth as primary health services, cleaner water supplies and sanitation reduce infant deaths.

Agricultural growth may begin with extensive development, as little-used lands – forest, bush, rough pasture, marshes etc. – are brought into cultivation as fields or managed pastures. Output grows, even if yields per unit area increase little, if at all. While land with agricultural potential exists, this process will predominate over intensification of production since it usually costs less to bring new land into production than to raise yields on existing farmland (Boserup, 1963; Pingali et al., 1987).

Once the agricultural frontier closes, however, agriculture grows largely through intensification. More crops a year can be produced from farmland by reducing fallow times, and by increasing the number of harvests a year – irrigation can make the latter possible even when rain falls only in one season. Yields per hectare can be raised by applying more manure and fertiliser; by more careful weeding – and in the process using more labour; and by investing in new technology embodied in improved seed, manufactured fertiliser, agro-chemicals, veterinary supplies, etc. (Nin-Pratt, 2015). The value of production per hectare tends to rise further as the composition of farm output, originally dominated by staples – cereals, roots, tubers and pulses – shifts towards higher-value items demanded by increasingly affluent urban households, including meat, fish, milk, oils and fats, fruit and vegetables.

As output increases and a surplus is generated, farmers engage more with markets for output and inputs, and often increasingly with labour markets – especially at peak times (Barrett et al., 2017).

3.3.1 Agricultural development usually promotes overall rural development As agriculture develops, the rest of the rural economy usually also does so, creating new sources of income for members of farm households and for some landless households. This allows households to diversify their sources of income, although individual members may come to specialise in particular activities.

38

Figure 15 Stages of agricultural development, after Timmer 1988

Box 1 Timmer’s stages of agricultural development

Timmer proposes four stages of agricultural development (see Figure 15).In the first, ‘Mosher’ stage, government invests in agriculture to start growth and

transformation, with two investments standing out in particular: roads to connect farms to market, and agricultural research and extension to generate improved technology.1 Once agriculture has started to grow (faster than the population growth), it enters a second, ‘Johnston-Mellor’ phase where it has to contribute considerably towards overall economic growth while releasing labour, capital and some land to other sectors.

It is these two early phases with which we are concerned in this framework. Two subsequent phases see agriculture more fully integrated with the rest of the economy in a

‘Schultz-Ruttan’ phase, as the rural labor force begins to shrink absolutely; and completely integrated in a ‘Johnson’ phase where agriculture is now a small share of both GDP and employment.

Policy settings:

1. Getting agriculture moving: Institutional change; new technologies; structure of markets and incentives; significant investments in rural infrastructure

2. Agriculture as contributor to growth: Establish market links with industry; technology and incentives to create a healthy agricultural sector; improve factor markets to mobilise rural resources

3. Integrating agriculture into the macro-economy: Declining share of food in urban budgets; push to make agriculture efficient via improved infrastructure and market linkages; shift resources out of agriculture, but substantial income distribution problems from lagging rural labor productivity Agriculture in industrial economies: Small share of food commodities in consumer budgets; income distribution is a political issue; unemployment in industrial sector creates pressures to keep labour in agriculture; environmental concerns and way of life issues

Source: Timmer, 1988; 2009

1 Mosher distinguished between fundamental conditions for agricultural development — markets, price incentives; technology including inputs and equipment; and accelerators — credit, education, land improvement, planning, and collective action by farmers.

39

The RNFE6 tends to develop in parallel with agriculture. In particular, increased agricultural productivity and output are associated with increased off-farm value addition, as Timmer highlights in his second stage. Empirically, rural household participation in the RNFE tends to rise with national income (Haggblade et al. 2007). Two key factors drive this: complementarity in public policies and investments in rural areas; and, linkages from agriculture to the RNFE.

Complementarity of policy and investments. Many of the public measures to stimulate agriculture – such as ensuring an enabling rural investment climate, establishing basic economic institutions, construction of and maintenance of rural roads, provision of electricity, education, health and clean water, etc. – benefit all economic activity in rural areas, not just farming (Wiggins, 2014).7

Linkages from agriculture to the RNFE arise through demand: agricultural development creates linkages in the local economy. Increased farm production requires inputs and services to grow crops and subsequently to process, store and transport produce. Links also arise in consumption when farmers spend some of their earnings on locally-supplied goods and services – housing improvements, furniture, education, entertainment, travel to local towns, etc. Hence multipliers from growth of agriculture to that of the RNFE can be strong. Studies suggest that each dollar of additional value added in agriculture generates US$0.60–0.80 of additional RNFE income in Asia, and US$0.30–0.50 in Africa and Latin America (Haggblade et al., 2007).

6 According to Barrett, Reardon, and Webb (2001), farm (or agricultural sector) incomes are ‘derived from the production or gathering of unprocessed crops or livestock or forest or fish products from natural resources’ (primarily land). All other economic activities are non-farm (or non-agricultural), even if they involve the processing or transport of farm production. Off-farm activities are economic activities performed outside the property owned by the household, including wage labour on a commercial farm or in a rural or urban non-farm enterprise.

7 The overlap between public investments for agriculture and those for the RNFE is considerable: indeed, almost all major investments potentially benefit both sectors. Items that only serve one sector, such as agricultural research and extension, typically represent only a small share of public spending in rural areas.

8 It costs more per capita to provide public services in more remote areas, and these areas tend to be poorer as well, making subsidies from urban areas necessary.

Not all RNFE activity is linked to agriculture. Some rural manufacturing may take place. Rural artisans may manufacture crafts and souvenirs whose value depends in part on their being produced in a particular place, perhaps using local raw materials, and embodying some craft tradition. Historically, putting-out systems have seen urban traders provide raw materials to rural workers skilled in spinning and weaving, collecting them when processed to yarn, cloth and garments. The system takes advantage of both low rural labour costs, and the seasonality of farm work. Industrial plants may move out of cities to avoid high rents, but generally they remain fairly nearby (less than one hour’s drive away): such plants are rarely seen in the deep countryside.

Rural manufacturing is vulnerable, however, to competition from low-cost factory-produced goods: plastic wares replace baskets and pottery, galvanised steel roofs replace local thatch. Hence rural manufacturing may well decline with development (Haggblade et al., 2007).

Economic growth usually permits an increasing range of public services to be supplied in rural areas – education, health, water and power supplies, police, etc. – giving rise to formal, salaried jobs. Although some of these may be financed from local taxes, national budgets support most of this activity.8

3.3.2 Drivers of changeSeveral factors drive these changes. Demand for agricultural output rises from urbanisation and rising incomes, with an increasing share of food budgets spent on higher-value produce rather than staples (Diao et al. 2003).

40

Some farmers may be able to produce for international markets, thereby overcoming limits to demand arising from the relatively small size of the national economy – the ‘vent for surplus’ case. Export opportunities tend to be restricted to crops that can be produced only from certain ecological niches, and which have high value for weight so that transport costs to distant markets can be tolerated (Poulton et al. 2008). Tea and coffee are good examples.9

On the supply side, imports or the growth of domestic industry can provide farmers with access to better and cheaper inputs: improved seed, veterinary medicines, manufactured fertiliser, irrigation pumps, two-wheeled tractors, etc.

Public policy and investment play a key role, especially in the early stages of agricultural development. Building and maintaining roads that link rural areas to cities and ports reduces transport costs, thereby raising prices for output, and reducing prices of inputs at the farm gate. Investments in rural electricity can be useful for operating processing machinery. Public investment in irrigation and drainage works can allow development of resources that require passing scale thresholds, as well as providing employment in the off-season for dryland farming.

Public agricultural research can generate technical improvements for local farming conditions. Late developing countries benefit from primary research already carried out in other countries and by international agricultural research centers, so that national systems can concentrate on adapting ideas to local circumstances. Public extension systems can help transfer innovations to farmers.

Government can further support agricultural development by investing in rural services – schools, health posts, clean water systems – for more educated and healthier rural populations.

9 It has been argued that as land and labour become scarce in emerging markets and in developed countries, SSA countries will be able to export a wider variety of crops (Filmer and Fox, 2014).

10 This is not to argue that agriculture may not be taxed at all. In the early stages of development, agriculture may be the only sector that is capable of generating public revenues to allow public investments. The point, however, is that such taxation should be moderate – and not the 50% plus implicit taxes that were previously applied to some export crops in Africa during the post-independence industrialisation period (Krueger et al., 1991).

Decentralisation of public administration may also create more formal employment in rural areas.

A necessary public role for public investment lies in assuring an enabling rural investment climate of law and order, a stable macro-economy, and establishing basic institutions – agricultural produce standards, land tenure, contract law, etc. – to reduce transaction costs and investment risk. Governments need to ensure that policies do not turn the terms of trade against agriculture (‘negative protection’) through heavy over-valuation of the currency, industrial protection, direct taxation of export crops, or operation of inefficient and monopsonistic marketing boards.10

Agricultural growth may be further stimulated if modern media and telecommunications can be used to transmit information on prices and technical innovations to farmers. Foreign direct investment in farming and the agricultural supply chains can both increase capital investment and bring in know-how. The development of rural financial services may help farmers to obtain credit for investment, and insurance against the risks of doing so. Remittances from rural migrants may also help farming households invest in improved production.

Many of these drivers also help business in the RNFE respond to increased demand for goods and services. Improved roads allow more interaction with towns and cities, thereby making it easier to obtain tools, machinery, inputs and financial services for non-farm business. Better roads also help those RNFE activities that produce for urban markets. Electricity supplies can power machinery, as well as allow workplaces, offices and shops to work in hours of darkness. Investments in rural people may have a stronger effect on RNFE business than agriculture, since literacy and numeracy often has higher returns in non-farm business than in agriculture. The same rural investment climate

41

that supports agricultural development also encourages investment in rural business.11

3.3.3 Making agricultural and rural development more inclusiveThe benefits of agricultural development will more likely to be widely spread, and especially among those on low incomes, under the following conditions:

• The initial distribution of access to land, with secure tenure, is relatively equitable so that most rural households can farm and herd (Box 2). Rights to land and other natural resources, especially those collectively held, also need to be secure against expropriation since those on low incomes often depend disproportionately on these resources.

• Technical advances are scale-neutral, so that even those with very small farms can benefit from them, and risk-averse farmers can try innovations on a trial plot before fully committing themselves to innovations. It also helps if technical improvements do not require much financial outlay, so that those with limited means can adopt.

• Improved techniques, or new farm activities, are intensive in labour either on the farm or in processing, thereby generating jobs both for farm households and for landless households.12

• Initiatives by private firms, farmers’ associations, non-government organisations (NGOs) or government find ways to overcome market failings for inputs and financial services that so often mean that those on low incomes cannot get access to these at reasonable cost. Those failings may arise through monopoly power in the supply chain, or through high transaction costs for farmers, input suppliers and bankers owing to information scarcity and asymmetries.

11 See Filmer and Fox (2014), Haggblade et al. (2007; 2010), and Wiggins (2014) for a discussion of policies which support the rural non-farm economy.

12 To some extent this will arise from factor proportions. In areas with abundant and under-employed labour, farmers are likely to adopt techniques and crops that use labour.

13 This is persistent to farmers the world over. When some farmers find a way to markedly increase their productivity, one consequence is often a surge of production and given the relatively inelastic demand for agricultural output, prices are pushed down. For those farmers with the productivity gains, their incomes may well still rise; but for farmers who have not seen such advances, the effect will be loss of income.

The RNFE is most likely to benefit from agricultural growth, all other things being equal, when small farms develop since smallholders are more likely to spend increased incomes on locally-produced goods and services: they are much less likely to be able to afford luxury consumption that involves urban goods and services.

A broad-based development of the RNFE is more likely when rural people are better educated and healthier, and when rural financial services allow would-be entrepreneurs with few assets to obtain credit for their start-ups.

3.3.4 Threats to inclusive agricultural developmentSeveral circumstances and processes, however, may result in less inclusive agricultural development, as follows:

• Concentration of land either initially or during transformation, combined with abundant (landless) rural labour supply may then allow farmers to pay very low wages and treat labour badly. Land concentration also encourages rent-seeking over entrepreneurship (Box 2).

• Traders may have market power allowing them to pay farmers low prices and extract rents from them.

• Farmers on low incomes may take on excessive debt to finance investments, then find that they cannot repay leaving them open to repossession of their land by banks, or to usurious interest rates from private moneylenders and onerous commitments to them – such as debt bondage.

• Farmers may face falling prices for their output owing to large increases in production from some other part of the country, or internationally, that reduces prices.13

42

Box 2 The importance of land tenure for growth with equity

Institutions of land ownership can greatly influence the inclusiveness of the transformation, above all the distribution and security of land rights.

Historically, major differences can be seen between land-tenure regimes where much of the land belongs to a relatively small group of landlords, so that many in the rural population are either landless and depend on working as labourers on medium and large-scale farms, or farm as tenants; and those where land is more equally distributed so that few if any of the rural population are landless (Johnston and Kilby, 1975).

The danger with the former is that landlords pursue rents to the detriment of entrepreneurship; while the rents extracted from tenant farmers reduce their incentives to invest and innovate since some of the rewards of their efforts may be appropriated by landlords.1 The two effects can thus combine to stymie investment and innovation, and thereby restrict agricultural growth.

Not for nothing, then, some see land tenure as critical to the overall development of the economy and society (Johnston and Kilby, 1975; Moore, 1966). Indeed, path dependency may apply, so that tenure changes may condemn some countries with skewed distribution of land to decades of low growth (Mahoney, 2001).

Land policy has thus often been critical in the early development of countries with unequally distributed land. The challenge has been to establish land tenure that is socially equitable, while promoting efficient land use and stimulating investment and innovation in agriculture.

In some countries, land has been redistributed from landlords to the state or farmers. Variations include:

• nationalising the land, then organising production in large units (for example, USSR from late 1920s, China 1949–1977, Cuba since 1959, Ethiopia 1974–1991, Nicaragua 1979–1989) which has tended to disappoint in both subsequent growth of production and productivity. Economies of scale in agriculture tend to be limited beyond family-managed holdings (Wiggins et al., 2010);

• redistributing land from landlords to tenants and farmworkers (Japan, Taiwan, South Korea, Egypt, Peru). Such reforms can succeed, provided complementary policies support the families to produce are provided. (Mennen, 2017, Thiesenhusen, 1995); and

• assigning land ownership to the state or community, but with use rights allocated to households (China after 1977, much of Mexico from the 1930s until 2000). This is also the longstanding tenure found across much of sub-Saharan Africa.

Such tenure can, in the early stages of transformation, protect poor farmers from losing their land to richer households. However, in later stages it may encourage land fragmentation instead of consolidation and prevent rural families with smallholdings who prefer to work in the RNFE or migrate from selling part, or all, of their land.

More common have been attempts to regulate tenancy, by controlling the level of land rents, or crop shares, and by giving tenants more security of tenure.

Since the early 1990s, concerns over land have tended to focus on providing existing users, usually smallholders, with greater security of tenure, rather than redistributing land. More secure tenure should encourage farmers to invest and conserve their land, allow them to use land

1 This argument can apply particularly to sharecropping, since the implicit rent rises with production. Fixed rents should have no disincentive for the tenant. In practice, any distinctive inherent in sharecropping may be more than outweighed if landlords provide inputs on credit; while fixed rents may not be fixed for long if the landlord is effectively a monopoly supplier of land.

43

• Failure to manage collective resources results in depletion or degradation of the resource. Common examples include overdrawing of groundwater for irrigation, and overfishing at sea. While these problems affect everyone using the resource, those on low incomes may be particularly vulnerable to losses and increased costs.14

For rural non-farm businesses, growth may be less inclusive when:

• access to urban areas is easy and rural businesses face strong competition from those based in towns and cities

• failings in rural financial services make it hard for small business owners to start or expand their enterprises.

In sum, while for the sake of being relatively comprehensive, there is a long list of factors that make for more inclusive, or exclusive, agricultural and rural transformations, it is also clear that a few factors are especially important. Four stand out:

• relatively equal access to land and other natural resources

• building and maintaining rural roads to connect villages to district, regional and national centers

14 For example, when groundwater tables fall, richer farmers can afford to deepen their wells and install more powerful pumps; when fishing ground become less productive, richer operators may be able to buy improved gear that more effectively catch the remaining fish, or buy large boats that allow them to access more distant fishing grounds.

• provision of education, health services, water and sanitation so that rural youth have a decent start in life

• developing financial services so that all can have access to savings, credit and insurance.

3.4 Transformation of rural households

Changes in the rural economy are accompanied by changes to rural households.

In the rural economy, households are the smallest, yet most important, economic unit. In most cases they own or have use rights to a key factor of production, land, and supply labour for their own production or to others. Depending on their location and connectivity to the rest of the economy; their human, financial and physical capital; and the overall path of the structural and agricultural transformations, they can take advantage of new opportunities, or be left behind. An inclusive rural transformation leads to major improvements in the incomes and quality of life of rural households, including better health, nutrition, housing and utilities. Households also usually become more resilient to shocks.

Rural household transformation is characterised by increased incomes for household members engaged in agriculture either as farmers or labourers; and increased opportunities and incomes in the growing RNFE. In this way, rural,

as collateral to obtain formal credit, and enable them to transfer land to others – sharecrops, rents, sales – if and when the farm household has better options in non-farm activity.

When Mexico relaxed its prohibition on land sales by peasant farmers in who farmed under the ejido system, both farm and non-farm incomes went up as households were free to adjust their livelihood portfolio to their aspirations and opportunities (de Janvry et al., 2017).

In Africa, land-titling schemes providing secure tenure have also raised incomes as households spent less time defending their land from encroachment by others, allowing more time for higher-earning activities in the RNFE, and encouraging investments in soil fertility. This has particularly benefitted women, who appear to have been more vulnerable under less secure tenure schemes.

Source: Timmer, 1988; 2009

44

agricultural and structural transformation are linked; agricultural transformation increases farm incomes, while structural transformation increases incomes in non-farm sectors. Structural and agricultural transformation can thus combine to raise the welfare of rural populations (P. Timmer et al., 2012).

Three changes usually take place for rural households: in their use of labour; welfare; and in their composition and location.

Labour use. Initially, agricultural transformation may lead households to spend more hours on agricultural activities. More frequent cropping may reduce seasonal under-employment, while intensified operations demand longer hours spent on the farm. Labour demands on the farm can rise to outstrip household labour, so that even small farms begin to hire significant amounts of labour.

The growth of the RNFE may also affect labour allocation. Some household members may take up such opportunities and cease to provide farm labour.

Moreover, improved connections to cities and other parts of the country raise the prospects for migration and commuting, as household members find jobs beyond their village. Most migration is temporary or seasonal. Most migrants keep their links to the rural household, send remittances, and visit – including returning home to work at harvest time – and intend to come home at some point even if only on retirement.

Households may thus see their members work in more diverse activities, as individual members specialise in particular activities. Those households with farms and other businesses may increasingly hire labour to augment the workforce and replace members who have taken up other opportunities.

Welfare. Successful agricultural and rural development creates more jobs, and jobs with better returns. Incomes thus rise, leading to more

15 For rural youth, this can offset the drawback that most inherit less land and livestock than their grandparents received.

16 Drawing on data from DHS [World Bank, Gender statistics], for 33 out of 36 countries, the share of female-headed households had risen from the 1990s to the 2010s. In the recent years, typically between one quarter and one third of households have been headed by females.

consumption and the chance to build assets – in physical goods, land and livestock, and savings.

With more assets, and more diverse income sources, households become less vulnerable to shocks and more resilient to them when they occur. Not only does that reduce aversion to innovation that may entail risks, but also the reassurance to households has considerable intrinsic value – even if this is hard to measure.

Moreover, with economic growth, governments have more resources to provide services such as education, health, policing and utilities – clean water, electricity, drainage – for rural communities. Major improvements to the education and health of rural people have been seen across the developing world: in most cases today’s youth start their adult lives with much more human capital than their grandparents had.15

Composition and location of the household. In most developing countries, rural households have become smaller and also tending towards nuclear families. That is partly a consequence of fewer births as rural women’s fertility is falling; and partly from the break-up of large, extended family households. The latter have economies of scale and scope in labour, and offer members more security against shocks; but they come with considerable transaction costs, which means that when they are not necessary, people tend to break away into nuclear households.

In some countries, the share of households headed by females has been rising,16 owing mainly to men migrating.

As fertility falls and households shrink in size, the ratio of adult workers to children and elderly people rises – and correspondingly the dependency ratio falls. For a generation or so, until elderly dependants push up the ratio, the household will have a demographic dividend, in which workers’ earnings support fewer dependants so that either consumption can rise, or assets and savings can accumulate. This in itself can augment economic growth.

45

When migration occurs, rural households in effect become multi-locational, with a core in the home village, while other members live in towns, cities or other agricultural zones. The advent of mobile phones makes coordination of the household much easier, and in some cases also allows family members to make rapid and secure financial transfers.

Key indicators of rural household transformation include:

• declining rural poverty (measured by household consumption per capita)

• increased off-farm and non-farm incomes • reduced underemployment • increased household savings • improvements to human capital (health, education, nutrition).

3.4.1 Making household transformations more inclusiveSeveral factors and policies can make these changes more inclusive, as follows:

• Broad-based economic growth that is relatively labour-intensive creates the jobs that allows rural household members new options.

• Equitable inheritance, where household patrimony is evenly shared by offspring and the widowed spouse can help to ensure that all young household members start life with some assets.

• Public education, health services, clean water and sanitation are critical to ensuring that rural youth have enough education and good health to take up opportunities.

• Family planning services can be particularly important in allowing women to control their fertility. Lower fertility helps avoid the health risks of early and frequent pregnancies, while small households have more resources to invest in the human capital of each child.

• Services that give would-be migrants better information on job opportunities in distant places, that protect their entitlements to public services when they move, that combat trafficking and labour exploitation can help reduce the risks and raise the returns to migration. Conversely, policies to restrict rural migration are likely to raise risks,

costs, and leave migrants more vulnerable to exploitation.

• Because economic changes entail some risks, social protection can reduce the costs to the more vulnerable.

Finally, stewardship of the local environment, and adaptive management of common resources, matter for communities based on crops, livestock, fishing and forestry. Poorer households will have fewer options to manage the externalities of private resource use such as depletion of common water resources. Climate change is already creating more variable weather patterns, increasing risks for all rural households. To cope with these threats, countries are experimenting with institutional innovations which combine the flexibility of local control with national and international financing and expertise.

3.5 Gendered aspects of economic, agricultural, and rural transformationsMany of the main outcomes associated with economic transformation may have positive effects on the distribution of welfare and opportunities across society, for women as well as men. For example, if economic transformation results in reducing poverty, poor men and poor women should benefit. The same is true of other outcomes usually associated with a more productive economy – better governance or macroeconomic outcomes such as more sustainable growth.

However, it is well known that women and men behave differently in a variety of economic situations for several reasons, including differences in reproductive functions as well as in the social, behavioral and cultural expectations and norms associated with being a man or a woman (gender identity). These differences (biological and socio-cultural) affect how men and women behave toward one another and in society, they affect the power balance between men and women (individually, within the household and in the community), and they affect men’s and women’s ability to find and make use of economic opportunities to improve

46

their own welfare and the welfare of those they care about or for whom they are responsible.

Within the household, women and men have separate and joint interests which are balanced when decisions are made (Doss, 2013). Spouses do not necessarily pool resources, nor do they always work together to maximise household income. This means that policies that change women’s opportunities may affect bargaining power and, through this channel, household outcomes. Thus an individual woman’s behavior and response to opportunities is affected not only by factors specific to her, and the community and society she lives in, but also by relationships within her household. See Marcus and Harper (2014) for a discussion of how economic change may help effect a positive change in gender norms. The power dynamic in these relationships, and the extent to which women are able to articulate and exercise their own values and preferences, will affect the extent to which rural transformation improves the lives of rural women and girls.

Women are actors within rural transformation. As economic agents, women’s behavior is important to the welfare of the rural household and to inclusive rural transformation. Women provide a large share of the labour in agriculture. They also provide most of the labour in the unpaid care economy in the household and often the wider family. But lack of capacity – because of, for example, reduced access to education, compounded by norms and social structures that have positioned women as subordinate to men in many aspects of their lives – resulting in lack of time among other things, have reduced women’s ability to benefit from the changes and opportunities that structural, agricultural,

17 See Marcus and Harper (2014) for a discussion of how economic change may help effect a positive change in gender norms.

18 Definitions of women’s economic empowerment vary, but all include that women have the capacity and agency to participate in economic activity, and that they benefit from such participation. Oxfam has an extended definition: ‘Oxfam asserts that effective economic empowerment for women occurs when women enjoy their rights to control and benefit from resources, assets, income and their own time, and when they have the ability to manage risk and improve their economic status and wellbeing. However, for Women’s Economic Empowerment (WEE) to translate into meaningful empowerment, women must also have the autonomy and self-belief to make changes in their own lives, including having the agency and power to organise and influence decision making, while enjoying equal rights to men and freedom from violence.’ (Oxfam, 2017)

and rural transformation brings, as well as to contribute to advancing the transformation.

In many countries, however, rural transformation has indeed brought about changes in these norms and social structures that have been beneficial to women. Sometimes these changes happen because of economic necessity: for example, men migrate to urban areas, obliging women to take on more active roles in household production. Sometimes they happen because of women’s collective action, recognising the social nature of their constraints and working to increase their autonomy in attitude and behavior: for example, women’s self-help groups in South Asia and elsewhere.17

As a result, women’s economic empowerment is now widely accepted as an intrinsic objective, having social, moral, political, and economic benefits. A gender-inclusive rural transformation would increase women’s economic empowerment – their capacity to participate in and benefit from economic changes, in ways that respect their values and dignity – without reducing the same capacity in men and boys.18

In short, an agricultural and rural transformation is gender-inclusive if increased household income increases the consumption of all members, as well as their capacities and agency in all main dimensions of their life: economic, social, psychological, political and wellbeing.

Focusing on women and girls, an inclusive transformation:

• improves women’s economic outcomes and reduces gender gaps in outcomes derived from market economic activities

• protects married women’s rights to autonomous production decisions, including

47

the control and use of assets in case of divorce or widowhood

• improves the efficiency of the care economy, reducing women’s unpaid labour burden

• improves women’s and girls’ human development outcomes, including in nutrition, and reduces gender gaps

• increases women’s capacity to develop and act on their own choices in economic spheres, considering their joint and individual interests.19

Some might argue that an initial widening of gender gaps might be acceptable so long as women’s opportunities and welfare also improved. In the medium term however, gender gaps need to narrow for economic as well as social, moral and political reasons. Like economic transformation, achieving gender equality is a long-term process.20

A number of measures have been proposed to measure the gender dimension of rural household transformations. For instance, the Women’s Empowerment in Agriculture Index (WEAI) was developed in 2012 by a consortium of scholars and practitioners convened by IFPRI and OPHI, and tested in several low-income countries (Alkire et al., 2013). USAID funded the development and testing of the index for use in its Feed the Future agricultural development programme. The WEAI covers five domains:

• Production – individual or joint decision-making capacity and power about household agricultural production.

• Resources – decision-making power about the acquisition, use and disposition of household assets.

• Income – decision-making power about the use of household income.

• Leadership – membership of a community group, comfortable speaking in public.

19 As a practical matter, it may not be possible to separate out women’s capacity to develop and act on their own choices in economic spheres from their capacity to develop and act on choices in other spheres – e.g. social, political psychological. For example, if women fear for their safety if they try to take public transport, their economic choices are limited. See Fox and Romero (2016) for a discussion.

20 See UN (2016) for an elabouration of this point in economies at all levels of development.

21 Gendered indicators on empowerment are hard to find for low-income countries, especially in rural areas.

• Time – allocation of time to domestic tasks and satisfaction with time available for leisure activities.

This measure captures important aspects of rural women’s economic roles and the common constraints to maximising welfare such as the burden of household chores. However, it leaves out some significant roles or role changes which are affected by structural and rural transformation, the most important of which is women’s participation in the non-farm economy. Enabling factors for participation in this economic sphere, including access to financial services, and mobility, also need to be measured. Here, we have chosen to focus on the welfare of women and girls in four basic roles:

• as participants in the market economy (producers, labourers, and consumers)

• as mothers, wives and participants in the unpaid care economy

• as children, developing into adults • as citizens both in their rural community and in their country.

Indicators of inclusive gendered transformation are defined for each role, based on indicators more commonly available for developing countries.21 They include a gendered analysis of:

• time use (in market and non-market economic activity)

• type of income-generating activity (household production, off-farm, and non-farm)

• earnings per hour, self-defined underemployment

• human capital indicators disaggregated by gender

• attitudes toward female autonomy (capacity for agency)

48

• financial inclusion (as proxy for autonomous control of income).

Ideally, we would include measures of women’s decision-making power regarding the use of household income and the disposition of assets (as in the WEAI). While these indicators are measured in Demographic and Health Surveys (DHS), analysts have uncovered a number of inconsistencies in these indicators, which is why we omit them for now.22

It should be noted that in many countries, these indicators are only now being collected at in household surveys. For example, while DHS surveys now routinely collect data on attitudes toward female autonomy, these data were not collected in the past, so there are no quantitative time series to work with. In these cases, we have to rely on qualitative data.

3.5.1 Making individual’s transformations more inclusiveGiven the importance of context and the institutions that govern economic, political, social and psychological relationships, it is not easy to predetermine the drivers of gendered transformation.

Policies and programmes to support an inclusive economic transformation have to consider the social relations operating in the community and society, and how these influence economic outcomes. At the same time, women are clearly not a homogeneous group; other aspects of their identity, such as ethnicity, stage of life cycle and economic situation interact with gendered structures of constraints. An advantage such as economic class or education can in some contexts help women more effectively to realise their own objectives and aspirations as economic actors – producers, consumers and investors – but may not be effective in another context.

However, some constraints to women’s economic empowerment have shown up at some point in most economies, and therefore some drivers can be identified. In some cases these drivers are simply the gender-responsive design of

22 See Donald et al. (2016) for some examples of this. A seemingly innocuous question can have many different meanings in different languages, cultures and settings. This seems to be happening with the DHS questions on who in the family usually makes the decisions on a variety of expenditure categories such as food, health, large household purchases, etc.

public programmes (e.g. agricultural extension, health services); in some cases they involve prioritising investments that reduce women’s unpaid labour (e.g. water supply or child care); and in some cases the drivers are new legislation or enforcing existing legislation to provide equal rights for women to work, engage in market transactions, and participate in society.

Gender-responsive policies would aim to ensure that women have equal access to the knowledge, technology, financial services, land, and other assets present in the household or community which would improve their incomes or increase their welfare. Examples include:

• securing women’s control and use rights to land so that they can maximise earnings on plots they manage individually or jointly or can devote more time to working in the RNFE

• measures and programmes to help women control their fertility and public support for child care and early childhood development programmes, allowing women to work outside the house if they want to

• measures to support household and small businesses that are either operated by women, or which include women

• laws and regulations that prohibit male violence against and/or harassment of women, along with programmes to enforce them and ensure women’s safety in public places

• support for women’s associations and self-help groups which can make sure that rural women have access to the knowledge and services they need

• ensuring that public programmes in rural areas are gender-responsive, including ensuring access of women and girls to expanding social services

• providing safe water at or near rural households to reduce the time women and girls spend on obtaining clean water. Similarly, measures to improve access to energy, such as firewood, to reduce time taken to collect wood or other energy sources

49

• lack of (or removal of) legal obstacles to women owning assets (jointly or individually) and accessing financial services

• engaging with men and traditional leaders to raise awareness of gender and encourage changes to norms.

Processes which expand opportunities for men and women both within and outside their traditional spheres are more likely to be inclusive. An example would be the expansion of wage-earning activities in farm and non-farm sectors. Gender-based segregation at the industry, occupation and even firm level remains rampant even in rich, post-industrial economies, so it would be unrealistic to expect it not to occur as new industries and activities emerge in rural areas and secondary cities (Fox, 2015). But as various new activities develop, and eventually labour becomes scarce, these sectoral and occupational distinctions and divisions are more likely to collapse. Expansion of women’s economic opportunities outside agriculture tends to influence other areas of empowerment, including parents’ willingness to invest in girls’ education, and men’s willingness to purchase labour-saving assets for the household or engage with women about the use of household income.

Changed incentives in the rural economy that favour activities that women or men traditionally perform may not have gender-neutral effects. Indeed, if they upset traditional relationships, potential retaliation could make women worse off – within the social relations of the household and beyond. For example, if men see women’s increased earning power as a reason to reduce their contributions to household expenses or household chores, women may find themselves increasingly burdened by the time and mental energy demands of remunerated activities, household chores and childcare, and trying to stretch the household budget. In some cases, men have responded to attitude or behavior change in women with an increase in domestic violence against women (Fox, 2015). To the extent that elites and political leaders support positive norm changes triggered by economic changes, rural transformation processes tend to be more inclusive.

Governments have opportunities to promote inclusiveness as part of national and rural

policy change as agricultural transformation proceeds (Table 1). In the initial stages, the main policy settings relate to institutional change (laws providing women’s use and inheritance rights to assets and outlawing male violence against women) and ensuring that as public services expand to rural areas, women and girls have access to them. While both settings may be difficult to enforce equally, setting the policy objective of equal access sends a message regarding expectations. In addition, legal protection combined with greater economic opportunities can help women increase their role in household production, both for domestic consumption and sales in markets, and resource allocation. Finally, policies and programmes that help women control their fertility (including programmes aimed at men) are crucial in the early stages.

In the second stage (where most lower-middle-income countries are), policies which support increased productivity in the RNFE as well as migration of young women become more important. More economic opportunities and choices usually facilitate more economic empowerment. Urban areas tend to be less conservative and offer more economic social mobility, which is why migration has been such an important driver of change in the lives of rural women. Simply creating economic opportunities for women outside rural households and farms has helped increase their autonomy, even in very constrained settings such as rural South Asia.

For those who remain in rural areas, farm investment as well as children’s development require cash, they need women to start a business or enter wage work – instead of being locked into non-remunerated household chores.

Government policies and programmes designed to discourage and punish male harassment of and violence against women are usually important to protect women, since legislation alone is not enough. Encouraging women’s involvement in civic associations and political life often complements more specific measures to ensure their mobility.

In the third stage, inclusiveness policies are broadly similar in rural and urban areas. Rural households resemble their urban counterparts in most dimensions. The income gap has

50

narrowed, and social services have expanded so that schooling levels, health outcomes, and quality of life indicators such as running water and ownership of consumer durables converge. In some cases, rural women may be left with a heavy burden of care owing to male outmigration. But their daughters have many opportunities. Staying in rural areas is now often a positive choice, made if they can make a good living and help support their families.

3.6 Conclusion

The framework proposed sees changes to the lives of rural women as embedded in wider processes of economic growth and transformation at national, sector, household and individual levels. Changes at higher levels facilitate those at lower levels, even if they do not determine them.

Agricultural development is seen to proceed through stages, differing by the declining importance of agriculture and the rise of industry and services, and hence by the changing roles and imperatives for agriculture. Within this context people rural areas experience economic opportunities, potentially leading to increased incomes and welfare – and the extent to which they are included or not in such improvements. For rural women and girls, processes of economic development can allow them to develop their capabilities and enhance their

wellbeing; but only if processes are inclusive and promote gender equality. Public investments and policies can contribute strongly to make sure that processes of growth and development contribute to inclusion and gender equality.

Table 1 summarises this thinking. It sets out the economic changes expected in the three phases of agricultural development, the policies that can lead to a gender inclusive development, and the changes that may be expected in the lives of rural women and girls in the different phases.

It is evident that in this schema, lives for rural women and girls improve incrementally and sequentially. Full development of women’s capabilities and gender equality is not achieved in the early stages of development, even if important steps – such as secondary education for girls – can be taken early in the sequence.

In this chapter we have referred to processes as drivers of change, distinguishing between different levels of aggregation where change can be observed. We have identified the key indicators of progress along the way, also at different levels of aggregation, which we will attempt to track in the case studies. We have also listed key factors identified in histories of economic development which have blocked the path of inclusive transformation at different times in different countries. We examine these processes and factors in our case studies.

51

Table 1 How opportunities for women and girls expand with agricultural transformation

Stages Early: getting agriculture moving – putting underused factors to work

Middle: agriculture as a contributor to growth, with transformation of economy

Late: urban economy, agriculture relatively small

Economic change Early development sees an economy that starts out as primarily agrarian and rural, with many goods and services produced by rural households for their own subsistence, become more commercialised, urban, and diversified – with much more manufacturing and services.

Rural households tend to be large, heavily engaged in agriculture, forestry, fisheries. Increased productivity and lower transaction costs create an income surplus for farmers.

All on left, plus:

Closer links from urban to rural; strong demand from urban for farm produce, especially higher value.

Rural markets develop, including for labor and land, rural financial services develop.

RNFE becomes a substantial source of employment and rural incomes. Includes activities directly linked to agriculture in supply chains (processing, transport, storage, etc.), as well as those indirectly linked through farmers’ spending, and those independent of agriculture (public services, some manufacturing).

Sharp slowing of population growth as later marriage and contraception become widespread.

Rural households tend to become smaller, with lower dependency ratios creating a demographic dividend. Rural out-migration means some households have remittances and a multi-site household.

All of those to left, plus:

Rural economy now diversified, with a RNFE larger than agriculture.

Rural population begins to decline absolutely.

Heavy out-migration of the young – children and older people begin to make up much of rural population.

Some households may lose their rural ties completely as all the young leave, so that when the parents die, the farm and house are sold.

52

Stages Early: getting agriculture moving – putting underused factors to work

Middle: agriculture as a contributor to growth, with transformation of economy

Late: urban economy, agriculture relatively small

Key public investments and policies in favor of women and girls

Investment in rural infrastructure to link markets and promote knowledge transmission (may help change norms, widen opportunities).

Public provision of schools, health posts – including contraceptive services, clean water, sanitation.

Laws: for land rights, especially for divorced and widowed women; outlaw violence against women and girls (including genital cutting) – latter may be hard to enforce but sends a message.

Encouragement of later marriage, smaller families, girls’ secondary schooling.

Encouragement of women’s self-help groups and associations in rural areas.

All of those to left, plus:

Stronger measures to encourage girls’ secondary schooling: may include financial incentives – as in some conditional cash transfers (CCTs), better provision of toilets and bathrooms, etc.

Measures to assist and protect young women migrants to cities – urban security, street lights, public toilets, labor standards, identity cards, etc.

Foster rural financial services, including savings groups and micro-finance.

Support development of rural industries where women are likely to work/contribute (e.g. dairy; poultry and eggs; flower and horticulture commercial farms) through complementary investments, enabling environment.

Encourage women to participate actively in civic life (possibly through use of quotas).

All of those to left, plus:

Encouragement for rural women to choose careers and undertake further education or training – promotion of role models for rural adolescent girls.

Changes seen in lives of rural women and girls

Improved health: fewer mothers and infants die

Fertility may begin to fall: later marriage, longer birth spacing, use of contraception.

Girls go to school, becoming literate and numerate; some go on to secondary school

More productive demand for household labor in agriculture, from increase in area cultivated, or from more intensive cultivation, spurs households to find efficiencies in the care economy.

Changes on left continue, plus:

Most rural girls go to secondary school.

Some young rural women may migrate to towns and cities: assembly plants, domestic service, offices. This encourages households to invest in girls’ development.

More scope for entrepreneurial women to set up small businesses. Some amenities such as piped water ease some of the chores typically undertaken by women and girls; electric light, concrete floors, plastic utensils, improved stoves, etc. further reduce the load — and fewer children lessens time spent in childcare.

Changes on left continue, plus:

Danger that out-migration is dominated by young males, so that women stay in villages with heavy unpaid responsibilities for children and the elderly.

As some rural women now obtain tertiary education and training, local women obtain more public service posts in rural areas.

In general, many more opportunities for rural women to work for wages.

53

4 Changing lives of rural women and girls over the longer run: Egypt, Peru and Thailand

4.1 Introduction

In the early 1960s, Egypt and Thailand were largely agrarian economies with most of the population living in rural areas, and engaged either directly in farming or working in the agricultural supply chains. Both had low average incomes, with most of the population living in poverty and especially so in rural areas.

Peru at that time was more urbanised with a substantial industrial and services economy and had achieved middle-income status. Peru, however, had marked geographic and social inequalities: the highland Sierra remained heavily agrarian, with the great majority of the rural population living in poverty. Moreover, while Egypt and Thailand saw sustained economic growth in the 1970s and 1980s, Peru experienced considerable set-backs in those decades arising from political dislocations, macroeconomic chaos, and civil war. By 1990, Thailand had caught up with Peru in average incomes, while Egypt had edged closer to average incomes in Peru.

By the mid-2010s, the three economies had become industrialised with most people living in urban areas. Poverty had fallen notably, average

incomes had risen to upper-middle-income status for Peru and Thailand and lower-middle-income for Egypt. To a considerable extent, they had passed through the transformations of economic structure and demography envisaged as integral to economic growth and development.

Key statistics that summarise these changes appear in Table 2 below.

What, then, changed in the lives of rural women and girls during the more than 50 years reviewed? This chapter sets out the record of economic growth and transformation, agricultural and rural development, and changes to the lives of rural women and girls from the early 1960s to the mid-2010s. It begins with Thailand, the country with the strongest record of economic growth; accompanied by other changes that have transformed the lives of rural women and girls – at least, in some dimensions. It can be taken as an exemplar, although as the details of the case will show, not necessarily a perfect one. Peru then follows, where several features similar to those of Thailand can be seen since 1990. Egypt’s story is less encouraging, with advances for rural women and girls hindered by gender norms and a labour market with abundant supply.

54

Table 2 Key statistics for development of Egypt, Peru and Thailand, 1965, 1990 and 2015

Statistic Year Egypt Peru Thailand

GDP per person, US$ constant 2010 1965 764 3,075 695

1990 1,588 2,696 2,503

2015 2,707 5,974 5,775

Increase 1965 to 2015, simple multiple 3.5 1.9 8.3

GDP share, agriculture 1965 29 18 32

1990 19 9 12

2014 11 7 10

Change, 1965 to 2015, percentage points (18) (11) (21)

GDP share, industry 1965 27 29 23

1990 29 29 37

2014 39 35 37

Change, 1965 to 2015, percentage points 12 6 14

Population growth rate, annual 1965 2.59 2.90 3.05

1990 2.47 2.14 1.38

2015 2.15 1.30 0.34

2015 rate as fraction of 1965 rate 0.83 0.45 0.11

Value of agricultural output, gross, per person US$ constant 2004/06 [3YMA]

1965 137 136 157

1990 147 114 220

2012 210 238 338

Increase 1965 to 2012, simple multiple 1.53 1.75 2.15

Land productivity: value of output per hectare of arable and permanent crop land, US$ constant 04/06

1965 1,590 585 373

1990 3,155 685 608

2012 4,861 1,295 1,390

Increase 1965 to 2012, simple multiple 3.06 2.21 3.73

Labor productivity in agriculture: value of output per agricultural worker, US$ constant, 04/06

1965 1,132 – –

1990 1,525 – 561

2012 2,704 2,409 1,059

Increase 1965 to 2012, simple multiple 2.39 – –

Under-five mortality rate, per 1000 live births 1965 267 192 123

1990 86 80 37

2015 24.0 17.0 13.5

2015 rate as fraction of 1965 rate 0.09 0.09 0.11

Fertility rate 1965 6.4 6.9 6.1

1990 4.7 3.8 2.1

2015 3.3 2.4 1.5

55

Sources: compiled from data series in FAOSTAT and World Development Indicators

Statistic Year Egypt Peru Thailand

2015 rate as fraction of 1965 rate 0.52 0.35 0.25

Contraceptive use prevalence by adult married women 1970 25 26 15

1990 48 59 66

2012 58.5 75.5 79.0

Increase 1970 to 2012, simple multiple 2.34 2.90 5.27

Primary school enrolment, adj. net 1970 64 76.5 76

1990 94

2015 99 98 91

Increase 1970 to 2015, simple multiple 1.55 1.28 1.20

Secondary school enrolment, net 1970 26 – 16

1990 86 26 –

2015 82 78 83

Increase 1970 to 2015, simple multiple 3.15 – 5.19

Age at first marriage 1970 – – 22

1972 – 22 –

1976 21 – –

1990 – – 24

1992 22 24 –

2010 – 23 25

2014 22 – –

Increase, 1970s to 2010s, years 1 1 3

Gender Development Index 2000 0.806 0.929 0.978

2010 0.851 0.952 0.998

2015 0.884 0.959 1.001

Increase 2000 to 2015, % 9.7% 3.2% 2.3%

Gender Inequality Index 1995 0.665 0.547 0.400

2010 n/a 0.406 0.319

2015 0.575 0.385 0.366

Decrease 2000 to 2015, % - 13.5% - 30.0% - 8.5%

56

4.2 Thailand: an imperfect exemplar

4.2.1 BackgroundThailand is unusual in South East Asia in never being colonised. Nevertheless, to preserve its independence in the nineteenth century, the kingdom had to fall in line with demands from European powers. In 1855 the kingdom signed the Bowring treaty with the UK that brought in free trade, with British supervision of economic policy. These measures established Thailand’s position within the global economy: an exporter of primary products, with rice, rubber, teak and tin to the fore; an importer of manufactures (Dixon, 1999; Jansen, 2001).

That pattern largely persisted until the 1960s. The economy in 1960 was largely agrarian: agriculture employed 80% of the labour force and contributed 37% of GDP. Incomes were low (Table 2) with 57% of the whole population and 61% of the rural population, living in poverty in 1962/63 (Krongkaew, 1985).

4.2.2 Development and transformation

Economic development In the 1950s, Thailand had tried to stimulate a more diversified economy through state enterprises in manufacturing. The disappointments of this model led to rethinking in which the state decided to restrict itself to investing in infrastructure and providing services, while leaving the private sector to take care of business. This required a political compact, since much private enterprise was in the hands of people of Chinese origin. Under the compact, they would be free to operate their businesses, so long as they left political leadership to the military and bureaucracy.

In 1961 the first national plan was published, which aimed to stimulate industry and transform the Thai economy. In marked contrast to the state direction of industrialisation seen in East Asia, Thailand opted for a laissez-faire industrial strategy.

During the 1960s and 1970s industry developed mainly based on processing agricultural exports. In the 1980s industrialisation was further stimulated by foreign direct investment (FDI) in manufacturing,

as Japanese companies facing rising domestic labour costs sought factory locations where labour was cheaper. In the 1990s investments in real estate and finance added to the mix.

In 1997/98, after rapid economic growth, a financial crisis erupted owing to the over-valuation of the Thai baht and a credit bubble. The baht was floated, rapidly falling to half its former value against the US dollar. Many firms and some finance houses went bankrupt, workers were laid off, and the economy went into recession. Reforms, however, were effective and within a few years the economy had recovered its growth; the kingdom paid off a bail-out from the IMF, and the baht appreciated.

In just over 50 years, GDP per person, in real terms, rose by more than eight times. Economic growth was driven by the following factors:

• policies and macroeconomic management that for most of the period encouraged private investment

• public investments in roads, irrigation schemes, health and education

• the presence of a large Chinese ethnic minority, well integrated into Thai society, with entrepreneurial ethics and skills. Much of formal private enterprise, including business in the agricultural supply chains, was in their hands

• during the 1960s and 1970s, aid from the United States, designed to prevent Thailand from becoming the next ‘domino’ to fall in the region, funded the construction of a road network that reached deep into the rural areas.

In the process, the economic structure was transformed (Figure 17). Agriculture, which in 1961 made up 36% of GDP, was less than 10% by 1993. Industry’s share of the economy rose from 19% in 1961 to 39% by 1991.

Demographic transitionSince the 1960s, Thailand has undergone a remarkable demographic transition. Population was growing at more than 3% a year in the early 1960s: by 1991 that had fallen to 1% a year, and to 0.3% a year by 2015. (Figure 18) The urban share of population has risen from less than 20% in 1960 to 50% by 2015.

57

Given movement from rural to urban areas, rural population growth that was 2.8% a year in 1960 slowed down so much that since 2000 the rural population has fallen, from 43 million in 2000, to just 33 million persons, in 2015.

Population growth has slowed as the fertility rate has fallen (Figure 19) from an average of

more than six per woman in her reproductive years in the early 1960s to 1.5 in 2014. At such a low rate, Thailand’s population will decline, once the current large cohorts from ages 30 to 50 enter old age.

Although rural fertility rates have been higher than the national rates, they have fallen faster

Figure 16 Economic growth, Thailand, 1961–2015

Source: Compiled from World Development Indicators

-15%

-10%

-5%

0%

5%

10%

15%

0

1,000

2,000

3,000

4,000

5,000

6,000

1961

1963

1965

1967

1969

1971

1973

1975

1977

1979

1981

1983

1985

1987

1989

1991

1993

1995

1997

1999

2001

2003

2005

2007

2009

2011

2013

2015

Annu

al g

row

th (G

DP p

er c

apita

) GDP per capita (US$ constant, 2010)

Figure 17 Structure of production, Thailand, 1961–2014

Source: Compiled from World Development Indicators

0%

10%

20%

30%

40%

50%

60%

1961 1966 1971 1976 1981 1986 1991 1996 2001 2006 2011

Shar

e of

GDP

Agriculture, value added Services, etc., value added Industry, value added

58

than urban rates. In 2012, the rural rate was measured at 2.1, compared to 1.5 in urban areas (MICS, 2012).

Fertility has declined as women took up contraception (Figure 20) from 1970 onwards, with usage rising from less than 15% of married women to more than 60% within a decade. In

the 2010s the proportion has risen to almost 80%: by 2012, rural use was slightly higher than urban rates, at 80%.

Much of the credit for rapid increase in uptake of contraception in the 1970s lies with the model forged by the National Family Planning Programme in the early 1970s that

Figure 18 Population growth, Thailand, 1961–2015

Source: Compiled from World Development Indicators

0.0%

0.5%

1.0%

1.5%

2.0%

2.5%

3.0%

3.5%

0

20

30

40

50

60

70

80

1961

1963

1965

1967

1969

1971

1973

1975

1977

1979

1981

1983

1985

1987

1989

1991

1993

1995

1997

1999

2001

2003

2005

2007

2009

2011

2013

2015

Annual growth

Mill

ions

Urban Rural Annual growth rate

Figure 19 Fertility, Thailand, 1961–2014

Source: Compiled from World Development Indicators

1

2

3

4

Fert

ility

rate

5

6

7

1961

1966

1971

1976

1981

1986

1991

1996

2001

2006

2011

Figure 20 Women’s use of contraception, Thailand, 1970–2012

Source: Compiled from World Development Indicators; MICS, 2012

10%

30%

50%

70%

90%

1970

1974

1978

1982

1986

1990

1994

1998

2002

2006

2010

Cont

race

ptiv

e pr

eval

ence

(any

met

hod)

, w

omen

age

s 15

–49

59

led to village-level services provided by local paramedics (Box 3).

Agricultural development Since the early 1960s, agricultural growth has averaged more than 3% a year, so that by 2013 the value of output, in constant terms, had grown by more than five times: well ahead of population growth (Figure 21).

Thailand’s agricultural growth can be divided into two phases: before and after 1990. In the initial phase, growth came largely from bringing additional land, often forest, into cultivation, taking advantage of the relative abundance of land with agricultural potential in the kingdom. With a growing rural population, the rural workforce expanded at the same time. Consequently, growth was extensive and productivity of both land and labour increased only a little during this period (Figure 22).

After 1990, the land frontier began to close – not least because forests were more effectively protected from agricultural encroachment. At the same time the slowing growth of the rural population, plus the competing attractions of

non-farm jobs and migration from the villages, meant that labour began to leave agriculture. Growth thus became more intensive, with higher yields per hectare and per worker.

Agricultural growth has not only been sufficient to outstrip population growth but has also allowed Thailand to export ever-greater amounts to world markets (Figure 23). New crops and products have been added to the list of exports: cassava and sugar in the 1970s, later fruit and vegetables, chicken, shrimp and tuna. Moreover, increasingly exports have been processed: canned chicken, tuna; pet food; and beverages (Dixon, 1999).

Agricultural growth has resulted from several factors, as follows:

• Thailand’s endowment of land suitable for cropping, which allowed farmers to expand the cultivated area, above all before 1990.

• Traders, processors and exporters in the agricultural supply chains who have effectively linked smallholders to markets, often providing inputs and in some cases credit to them.

Box 3 Thailand’s highly successful family planning program

The National Family Planning Program began in 1967. Initially, this had limited success, but during the early 1970s an effective program was devised through the combined efforts of an NGO, the Population and Community Development Association (PDA), led by a determined and charismatic former economist who had headed the National Economic Development Board, Mechai Viravaidya, and the Ministry of Health. Mechai became nationally known through his highly visible promotion of condoms that aimed to take away people’s embarrassment.

But Mechai was more than a showman: together with health ministry officials, he trialed ways of reaching rural women with messages about family planning and supply of contraceptives. The breakthrough came in 1974 when the government authorised village shopkeepers to distribute contraceptive pills: formerly this could be done only by doctors. Later village nurses were trained to administer injectable contraceptives (Depo-Provera) and carry out some forms of sterilisation.

The key was providing services in the villages, run by local people whom women knew and could trust.

The PDA backed up its family planning with rural development, including water jars for rainwater catchment, toilets, loans of pigs, rice and buffaloes, and revolving loan funds. Participation in family planning meant more access to loan funds, although the PDA claims that no one was penalised for not practising family planning.

Sources: Drake et al. (2014), Frazer (1992), Jitamontree (2007), Korten (1980), Rosenfield and Min (2007), WHO (2004)

60

• Public policy and investment. The role of the state has not necessarily been perfect. In the 1960s and 1970s, for example, rice was taxed and sometimes exports were restricted to keep domestic prices low. In 1955 the rice taxes amounted to 58% of the wholesale price (Rock, 2002). Rice production increased

despite such taxes, largely because the natural conditions for growing rice in the central plains were so good that costs of production were low and farmers could still make a living despite repression of output prices.

• Moreover, for much the period 1960 to 1990, state investments in agriculture were

Figure 21 Agricultural growth, Thailand, 1961–2013, three year moving average

Source: Compiled from FAOSTAT data

0%

1%

2%

3%

4%

5%

6%

7%

0

5

10

15

20

25

30

35

1962

1964

1966

1968

1970

1972

1974

1976

1978

1980

1982

1984

1986

1988

1990

1992

1994

1996

1998

2000

2002

2004

2006

2008

2010

2012

Annual growth rate

Valu

e ag

ricul

tura

l out

put

(US$

bill

ion

cons

tant

, 200

4/06

)

average: 3.36% per year

Figure 22 Agricultural productivity, Thailand, 1961–2013

Source: Compiled from FAOSTAT data

200

400

600

800

1,000

1,200

1,400

1961

1963

1965

1967

1969

1971

1973

1975

1977

1979

1981

1983

1985

1987

1989

1991

1993

1995

1997

1999

2001

2003

2005

2007

2009

2011

2013

US$

cons

tant

200

4/06

per

hec

atre

, or

agr

icul

tura

l em

ploy

ee

Prod Value/Agri LabValue prod/ha

61

Figure 23 Agricultural exports, Thailand, 1961–2013

Note: Exports include rubber, rice, sugar, cassava, palm oil, pineapple, canned chicken, prepared food, non-alcoholic beverag-es and pet food. Not shown: seafood (tuna and shrimp) = US$7 billion by 2013.Source: Compiled from FAOSTAT data

0

5

10

15

20

25

30

Exports

1961

1963

1965

1967

1969

1971

1973

1975

1977

1979

1981

1983

1985

1987

1989

1991

1993

1995

1997

1999

2001

2003

2005

2007

2009

2011

2013

US$

billi

ons

(con

stan

t, 20

02)

Imports

modest, largely restricted to building roads, constructing some irrigation works23 and agricultural research. Despite the many government programmes for agriculture, the tended to be centralised, uncoordinated and had had little impact. Instead, it seems that farmers took the initiative, facilitated by local traders. The government has followed behind. (Dixon, 1999)

The combination of these factors has produced one of the more remarkable examples of agricultural development of the second half of the twentieth century: the conversion of the lagging Northeast, long regarded as an agricultural backwater, into a powerhouse for agricultural exports (Box 4).

23 Around 38% Thailand’s arable area was irrigated in 2013. Repeatedly government has promoted visions for major increases in irrigation, based around large dams and canal systems. Few of these schemes have been implemented, partly because more detailed planning has shown the works to be more costly and difficult, and water less plentiful, than originally imagined (Molle et al., 2009). Nevertheless, the irrigated area increased from 1.6M to 6.4M hectares between 1961 and 2013.

24 Some ideas have failed: village crafts have been hit when the price of finished silk products fell, gem cutters left the area, and hand processing of cassava, cashew and sesame has not been able to compete with industrial units.

Rural diversification: the rural non-farm economy and migrationAgricultural growth has been accompanied by strong growth of the RNFE. Employment in the RNFE has grown to reach around half of all rural jobs. Some activity is linked to agriculture, such as rice mills, sugar refineries, workshops manufacturing and repairing agricultural machinery, and transport services. Other local businesses have bene established to meet the demand from farmers for goods and services.

In addition, some rural manufacturing – other than agricultural processing – has grown. Examples include processing local raw materials, for example, cutting gems, weaving silk and making artificial flowers.24 Other artisan manufacturing has developed through a putting-out system where merchants from Bangkok provide village workshops with materials from which to produce handicrafts. Some larger-scale, more formal industry has

62

Box 4 Transforming the Northeast from backwater to export powerhouse

Before the 1970s, Northeast Thailand, or Isaan in Thai, had long been neglected. Remote from Bangkok with a climate prone to droughts, little irrigation and otherwise modest natural resources, it was the least developed part of Thailand. Indeed, the rural population of Isaan was the butt of national jokes about country bumpkins.

Since the 1960s, however, the regional economy has been transformed to become Thailand’s fastest growing region, its economy driven by exports of cassava, rubber, sugar cane, grapes and vegetables. Average incomes tripled, in real terms, between 1970 and 2004.

Change began with agriculture. Production of the main staple, rice, which still occupies around 70% of the land, doubled between 1973 and 2003. Growth was mainly extensive in the 1970s and 1980s as forest was cleared to create new fields. The rice-producing area expanded: yields rose only modestly from around 2.5 tonnes per hectare in the late 1970s to just over 3 tonnes in the early 2000s. With more land cleared, cash crops were added: cassava – for export, initially as chips for animal feed to Europe; maize – again a livestock feed; sugar cane; fruit and vegetables; and rubber. Almost all these are produced at low unit cost, as low as anywhere in the world, thereby allowing exports at a profit.

What has made the difference to the region? Public investment began with roads in the 1960s, when US money paid for roads into the

remoter parts of Thailand to counter the threat of insurgency at the time of the Vietnam war. Roads linked Isaan to the rest of Thailand and to the ports, making exporting possible. Irrigation works were constructed, although even today the region has little more than 17% of the cultivated area under irrigation. The government also invested in people, their education and health.

Traders have played a key role in linking smallholders to the rest of the supply chain. Ethnic Chinese traders had long been active in Isaan, providing, among other things, consumer goods and farm tools; but from the 1960s with the roads in place they saw commercial opportunities. They not only bought up produce, but made sure that farmers had the inputs to expand production. They built rice mills and warehouses. In recent times, traders and agri-businesses in the supply chains created have proved adept at ensuring that Thai farm exports meet the demanding standards of high-value markets, including that of Japan.

In the 1960s, the government set up the Bank for Agriculture and Agricultural Cooperatives (BAAC) to provide finance to smallholders. Unlike so many other state agricultural banks, BAAC has not only operated with only small state support, but also has successfully reached many smallholders, using group liability lending.

Other than the investments in roads, education and health, the government has tended to facilitate rather than lead this development. It has funded agricultural research to overcome some of the limits of Isaan’s modest soils and climate. Where links between farmers and agri-business have needed a kick-start, the government has intervened occasionally and judiciously in supply chains.

Not all change in the Northeast has been the result of local development. As once-lagging region, many of the young migrated to Bangkok and the coastal assembly plants that offered better returns than farming or other local employment (see the next section).

Sources: Ekasingh et al. (2007), Rigg et al. (2012), Rigg et al. (2014)

63

located in rural when assembly plants have decentralised. In this respect, some NGOs, such as the Population and Community Development Association (PDA), have actively attracted investors to rural locations. The PDA has, for example, helped create an industrial park in the Northeast for factories producing garments and footwear, mainly employing young women. It is not only jobs in the factories that count, but also the ancillary activity that accompanies them, in transport (including servicing the ubiquitous motorbikes), providing lunches to factory workers, and in shops and services where workers spend their wages.

Overall, non-farm rural jobs in Thailand tend to be more productive than those on farms, even if less than those in urban areas: in 1999 labour productivity in agricultural work was estimated to be about one third the national average, while for rural non-farm activity the average was 80% of the national level (DAN, 2003).

Growth of the Thai economy since the 1960s has led to migration from rural areas to Bangkok and other urban centers, to work in assembly plants, construction, domestic service and other jobs such as retailing. By 1995, it was estimated

that 22% of the population had moved in the previous five years (Chamratrithirong et al., 1995): a 2015/16 migration survey reported that, between 2012 and 2015, 14% to 21% of the population had moved in the previous year (Laosutsaen, 2017). Among rural households with low incomes, most have at least one migrant member (Mills, 2001).

Migrants from rural areas are primarily young adults. Young men travel not only within Thailand, but also some move to the Gulf and wealthy parts of South East Asia such as Singapore, with construction being one of the main activities. Young rural women migrate at least as commonly as men, if not more so. Women form the majority of workers in manufacturing plants, as well as almost all those in domestic service (Garip and Curran, 2010; Mills, 2001).

While migrants may move for years at a time, most of them return eventually to their home villages, if only on retirement (Rigg et al., 2014)

Millions of rural Thais leave home every year to engage with work in other places; they leave behind their spouses, their parents, and their children. They also leave their villages and their land. What these migrants rarely do, however, is

Figure 24 Map of Northeast Thailand

Source: Wikimedia, Heinrich Damm, Aachen

64

desert their families, abandon their homes, or sell their lands (Rigg, 2019).

The rural transition: labour shortage and poverty reduction

Growth has generated jobs, especially informal ones. The periods of economic growth (1992-1996 and 2000-2002) boosted labour demand substantially in the sectors where the poor are a dominant group. As a result, wages and salaries of the bottom poorest groups increased dramatically, so did farm income and income in-kind. (Krongkaew et al., 2006)

With slowing population growth, out-migration, and more non-farm work, the farm labour force ceased to grow from around 1990 and thereafter declined (Figure 25) (Rigg, 2019).

25 Labour shortage has also encouraged immigration from neighbouring countries, most of which – Malaysia excepted – are poorer than Thailand. ‘In 2012, there were 110,000 migrants registered from the Lao PDR, 267,000 from Cambodia, and 1,111,000 from Myanmar’ (Huguet et al., 2012). ‘Irregular migrants are thought to outnumber regular migrants giving a total migrant labour stream of some 3.0 – 4.0 million, or approaching 10% of the Thai labour force’ (Rigg, 2019).

26 Estimates of rural poverty have been made for earlier years, but they are not necessarily compatible with the data series shown. An earlier series, for example, shows rural poverty from 61% in 1961 to 34% in 1990 (Jansen and Khannabha 2009).

Labour shortage has led to mechanisation of some farm operations (Isvilanonda et al. 2001: Funahashi 1996). Between 1990 and 2002 the stock of tractors – mainly two-wheel, small units – rose from just over 50,000 to almost 700,000 (FAOSTAT data).25

Competition for labour from manufacturing and services not only led to a reduction in the farm labour force, but also put pressure on agricultural wages. Real agricultural wages tripled between 1985 and 2003 (Bryant and Gray, 2005). They have almost certainly risen considerably since: wages of skilled agricultural workers rose by another 50% in constant terms from 2003 to 2013 (Wiggins and Keats, 2014).

As wages have risen, poverty has fallen. In 1988 almost three-quarters of the rural population were estimated to live below the national poverty line;26 by 2013 that had fallen to less than 14%. The gap between urban and

Figure 25 Employment by sector, Thailand, 1960–2011

Source: Groningen productivity database (Timmer et al., 2015)

0

10

20

Num

ber o

f em

ploy

ees

(thou

sand

s)

30

40

50

1960

1962

1964

1966

1968

1970

1972

1974

1976

1978

1980

1982

1984

1986

1988

1990

1992

1994

1996

1998

2000

2002

2004

2006

2008

2010

Agriculture Manufacturing Construction, mining, utilities Services

65

rural rates of poverty has narrowed considerably, from 30 to just 7 percentage points (Figure 26).

Higher incomes and welfare are readily apparent, even in the in the villages of the Northeast:

… for the Northeastern region noted above, poverty has fallen dramatically, the fear of under-nutrition virtually banished, real incomes have risen, health and education profiles have improved over a broad front, access to a swathe of amenities has widened, and livelihoods have diversified dramatically. (Rigg et al., 2012)

Income inequality has also fallen (Figure 27). The Gini coefficient peaked in the early 1990s at 0.48: by 2013 this had come down to 0.38.

4.2.3 Changes to the lives of rural and women and girlsThis is the background to the lives of rural women and girls: an economy that has managed, with considerable success if not without qualifications, to achieve economic growth, structural transformation, rural development and poverty reduction. Agricultural and rural development have been key elements of the story.

Changes to the lives of rural women and girls derive from three main influences: the development model outlined above; direct provision of services in rural areas that have improved health, education and facilitated family planning; and gender norms. These are seen in four basic roles of rural women, set out in section 2.5, as economic participants; mothers, wives and carers; developing adults in their human development; and as citizens and agents.

Women as economic participantsRural women have benefited considerably from the development model. Diversification of the rural economy has created new options for women, either as migrants or as workers in rural enterprises. Nationally, a clear picture emerges of women leaving agriculture – much more so than men – and taking up jobs in industry and services (Figure 28).

The tightening of labour markets since the late 1980s also means that the returns to their efforts have tended to rise. (Figure 29) Women’s wages have increased for various categories, but the largest increases from 1991 to 2005, were for women farm workers whose wages rose by 79%. Moreover, the gap between men’s and women’s wages narrowed, although in 2005 a woman

Figure 26 Poverty, Thailand, 1988–2013

Source: Poverty and Equity database, The World Bank. Poverty headcounts are at nationally-defined poverty lines

0%

20%

40%

60%

80%

1988

1990

1992

1994

1996

1998

2000

2002

2004

2006

2008

2010

2012

Head

coun

t

Poverty, rural Poverty, urban

Figure 27 Income inequality, Thailand, 1981–2013

Source: Poverty and Equity database, The World Bank

35

40Gini

coe

ffici

ent 45

50

1981

1985

1989

1993

1997

2001

2005

2009

2013

66

working in agriculture would still on average receive only 87% of wages paid to men.

Women as mothers, wives and carersReduced fertility has meant fewer children to care for, while improvements to housing should have reduced time spent on domestic work.

Rural electrification was promoted from the early 1970s, when less than 10% of rural households had access. Coverage is now 99% (GNESD, 2018)The government has also invested heavily in rural water supplies and sanitation, so that from just 10% of the population having improved water supplies in 1973, by 2015 more than 95% of the rural population had improved water, on their premises (Juntopas and Naruchaikusol, 2011; MICS, 2016).27

Despite such improvements, it still seems that in rural Thailand women do far more domestic work than men. A national survey in 2001 reported that in rural areas, women spend an average 5.4 hours a day on domestic chores and child care, compared to 2.9 hours for men (Time Use Survey, 2001).

In addition, as parents age, the expectation is that a daughter will take care of them. Young

27 Water quality, however, remains a concern. One response has been to encourage rural households to make use of rainwater: now a source for more than 8 million rural households.

migrant women are expected to send remittances to support the household to a far greater extent than male migrants are (Mills, 2001; 2003).

Women’s human developmentPublic investments in health and education mean that rural women and girls are healthier and better educated than their mothers and grandmothers were.

For example, in health, life expectancy for women in Thailand has risen from 57 years in 1961 to 78 years in 2014 (Figure 30). Mortality among under-fives has fallen from 147 per 1000 live births in 1960 to just 13 in 2014 (Figure 31). In both cases, most of the gains were achieved before 1990.

In education, Thailand had a high rate of primary school enrolment in the 1960s and by 1990, enrolment was almost universal (Figure 32). Literacy rates are thus high: in 2012 98.6% of rural women were literate (MICS, 2012).

Before 1990, however, secondary schooling lagged behind primary schooling: only 30% of eligible children were enrolled in secondary school. The belated recognition of this deficit subsequently led to heavy public investment,

Figure 28 Women’s jobs, Thailand, 1980–2013

Source: Compiled from World Development Indicators

0

5

Mill

ions

10

15

201319901980

ServicesIndustryAgriculture

31

14

5

2

13

9

4

8

Figure 29 Women’s wages, Thailand, 1991–2005

Source: Statistics from Labor Force Surveys (LFS) of Thailand for years 1991-2007, presented in Adireksombat et al., 2010

0

20

40

60

Wag

e pe

r hou

r (Ba

ht, 2

007)

80

100

120

2005/071998/001991/93

ProfessionalClericalService/sales

Agri.worker

67

with up to 25% of the public budget being spent on education (Fan et al., 2004). Consequently, by 2013, 85% of boys and 89% of girls were in secondary school. By 2012, there was almost no difference in enrolment between urban and rural areas.

Girls now seem to get as much or more access than boys to education. Indeed, a 2009 survey indicated parents spending more on their daughters’ education than on their sons’. This may be because daughters are expected to care for elderly parents, so that migrant daughters

Figure 30 Life expectancy, Thailand, 1960–2014

Source: Compiled from World Development Indicators

50

55

60

65

70

75

80

1960

1964

1968

1972

1976

1980

1984

1988

1992

1996

2000

2004

2008

2012

Year

s (fe

mal

e)Figure 31 Under-five mortality, Thailand, 1960–2014

Source: Compiled from World Development Indicators

0

20

40

60

80

100

120

140

160

1960

1964

1968

1972

1976

1980

1984

1988

1992

1996

2000

2004

2008

2012

Unde

r-fiv

e m

orta

lity

rate

(per

1,0

00 li

ve b

irths

)

Figure 32 School enrolment in Thailand, 1971–2012

Source: Compiled from World Development Indicators

75

80

85

90

% (g

ross

)

% (g

ross

)95

100

105

1971

1978

1985

1992

1999

2006

2013

Primary school enrolment Secondary school enrolment

Female enrolment Male enrolment

0

20

40

60

80

100

1971

1978

1985

1992

1999

2006

2013

68

have proved more reliable in sending remittances than sons have (Wongmonta and Glewwe, 2016).

Women as citizens and agents, gender normsMost gender norms in Thailand accord women a secondary role in a society in which men are expected to lead and take key decisions. That said, the degree to which women have been relegated has not necessarily been that great. One small proof of that emerges in the 2015/16 MICS: less than 9% of men and women agreed that a man could justifiably assault his wife.

Indeed, some very different interpretations of women’s ideal roles can be found:

There are competing views about gender relations in contemporary Thailand. On the one hand, it is argued that women traditionally have had a high status, and the processes of economic and social transformation are eroding that status. On the other hand, Thai women are presented as having being oppressed by patriarchal institutions which are underpinned by Buddhism and the Sakdina State. In this view the transformation appears to offer the opportunity for women to escape from the traditional patriarchal relations and improve their status. (Promphakping, 2000)

Four norms stand out, regarding property, business, care and mobility.

Property. Most groups in rural Thailand have matrilineal inheritance and matrilocal residence upon marriage. Hence, when a young couple marry, not only do they initially live with the bride’s parents, but also land rights come from the bride’s family. The new husband brings any capital his family may have plus any livestock. This does not mean that his wife has control over those assets: on the contrary, the new husband tends to assume the role of head of household. But it does nuance any sense that assets belong exclusively to men or that women only have rights through marriage (Bryant and Gray, 2005; Ganjanapan, 1994).

Matrilocal residence preferences, combined with bilateral kinship and equal inheritance practices, also tend to give women access to supportive social networks and material resources throughout their adult lives (Mills, 2005).

Business. Buddhist ideals see handling money and business as rather undesirable: they detract from a focus on spirituality and nirvana. In rural society, this has meant that men have preferred to leave business and household finances to women. Hence, when opportunities to participate in rural non-farm enterprises have arisen, women have been reasonably well placed to take these up.

Caring. Rural women are expected to be prime carers of both children and aging parents. Not only does that mean they spend far more time than men on caring for family members, but that they feel obliged to remit money when working away from home, and to return home frequently to assist their parents.

Mobility. Perhaps most important of all, gender norms have not prevented many young rural women from migrating alone to take up employment opportunities. This is not to say that allowing young women to travel unsupervised is approved of. On the contrary, it contravenes some ideals; but not sufficiently strongly to prevent movement. The combination of the household need for additional earnings, plus youthful desire to see new places and new sights, has trumped the ideal of the dutiful daughter who remains modestly at home (Mills, 2001; 2016).

Consequently, rural youth seek work in Bangkok not only to fulfill economic commitments to family at home but also to enact deeply felt desires for the sophisticated forms and practices of modern Thai citizens. (Mills, 2001)

Gender norms are not set in stone: conflicts between ideals and their practical expression are a fact of life, becoming all the more pressing when the context changes as much as it has across rural Thailand in the last generation or two. Some of these tensions play out in the identity of young women migrants, and the way that myths and gossip develop to try to impose order and normality on novel and disruptive processes (Box 5).

69

The overall point is clear: while gender norms in Thailand may not favour women, they have not much prevented them from migrating or from engaging in business outside the home.

4.2.4 Conclusion: the power of labour marketsSeveral lessons can be drawn from the Thai case:

• Economic growth across sectors can create jobs and livelihoods that create new opportunities for rural women and girls.

• The success of the family planning programme that got underway in the 1970s not only eased rural women’s lives, but also slowed the growth of the labour supply, so that, by the 1990s, labour shortage began to show up in rural areas, driving up rural wages.

• Government’s role was limited, but probably essential. For most of the last 55 years it has managed the economy reasonably well – and learned in 1998 just how costly it can be when such management fails – and encouraged investors and entrepreneurs, including those

from abroad. It built rural roads, with significant US aid; invested in health, clean water, and education – albeit with a delay in Realising the importance of secondary schooling.

• Advances for rural women were made possible by gender norms that did not unduly restrict rural women from migrating and obtaining off-farm employment.

Little of this peculiar to Thailand. The relative peace and stability of the kingdom is clearly another critical factor. The political bargain that took shape and reached its full expression by the early 1960s has underpinned that stability, although the kingdom has experienced political crises during the last 55 years. Several coups, and rural insurgency in the 1960s and 1970s, have broken the peace. But such episodes have been brief.

Thailand has been relatively well endowed with agricultural land, forest and coastal waters, but lacks the very high value natural resources that countries with abundant minerals and hydrocarbons have.

Box 5 From structural transformation to individual blame: making sense of change in contemporary Asia

Migration for young women in Thailand can lead to conflicts over identity. While young women carry with them the ideal of the dutiful daughter, once at the destination – most often Bangkok – they are surrounded by and tempted by lifestyles that stress consumerism and entertainment, that seen as more modern and progressive than any contrasting rural norms. In practice, it seems most women migrants adopt dual identities. (Clausen, 2004)

Mary-Beth Mills (2016) argues that some people try to make sense of the dislocations and inequities of economic growth and development by blaming individuals for their behavior. Women, mainly young women, are usually the target.

Hence in Thailand gossip is used to try to discipline young migrant women by telling tales about the idealised ‘good migrant’. She works hard, consumes modestly, eschews romances, sends money home and puts her parents and family first. The fate of those who do otherwise then becomes the subject of morality tales.

For Thai men who migrate internationally, the fear that their wives will misspend remittances, or take a lover in their absence, form another set of tales, often found in the lyrics of popular music.

In Japan and Singapore concerns about late marriage and low fertility are expressed in disapproval of young women’s behavior.

Absent from such narratives are considerations of the structural realities of making a living in contemporary Asia that compel people to migrate, and to defer marriage. Such tales are welcome for employers as well: the modest young migrant in the assembly plant is less likely to join a union, complain, or demand better working conditions.

Sources: Clausen, 2004; Mills, 2016

70

Indeed, the more the details of the Thai model are clear, the more imperfect the exemplar appears. It seems to be a case where government and people got the fundamentals right for most of the time, did not make too many errors, and found ways to correct them when necessary.

The great majority of rural women are undoubtedly better off than their mothers and grandmothers; in their incomes, job possibilities, health, education and control over their fertility. Rural girls today have possibilities that their grandmothers could only have dreamed about.

In both development and the situation of rural women, the current situation is far from ideal. Indeed, a review of the problems of contemporary rural Thailand produces a long list of significant challenges (Rigg et al., 2012). The rural economy has rescued people from hunger and deep poverty, but it has yet to deliver consumer prosperity for most. Income inequality remains quite high: some people are not doing as well as they should. Thailand now has several million immigrants from neighbouring countries who are vulnerable to poor treatment, and indeed in the worst cases, trafficked as slave labour on fishing fleets. The state has provided some public goods, but has yet to create a welfare state that provides for the old, infirm and unfortunate. Some economic activity has led to pollution, loss of forest and biodiversity. The kingdom is arguably more democratic than it was in 1960, but the current government is a military junta: prolonged civilian rule, with competitive parties prepared to cede power in elections, has not yet been established.

Rural women may be better off than before, but they are still largely subordinate to men in an economy and society where women are still expected to be, above all, wives and mothers.

4.3 Peru: a delayed start

4.3.1 BackgroundPeru’s development has been profoundly influenced by the impact of Spanish conquest in the early sixteenth century. Within 50 years the Spanish created an economy and society the main features of which have persisted. Above all, the economy was designed to extract minerals and any other transportable wealth from Peru,

while providing the Spanish and criollo elite with lifestyles appropriate to that of the gentry. The indigenous population that survived conquest and disease were relegated to manual work: deprived of the best land, and coerced into working in the mines and on the estates of the gentry. This economy was deeply divided by class, ethnicity and language.

Moreover, the elites, living off the rents of their estates and mines, were not particularly entrepreneurial. When new opportunities to export became apparent from the mid-nineteenth century to serve the new demands of industrializing Europe and North America, facilitated by improved transport, those investing in new activities tended to be expatriates from the industrialised world. Despite the mineral wealth of the Andes, in the early years of the twentieth century, Peru was one of the poorer countries of Latin America, with incomes averaging 25% of those of Argentina.

Nevertheless, until the 1970s the economy grew moderately well, at around 2.6% per person a year on average, with few interruptions other than the early 1930s when commodity prices slumped. From the 1940s, growth based on agriculture, mining and trading was supplemented by the expansion of domestic industry, protected by tariff barriers.

Yet the social divisions created 400 years earlier persisted. While the middle and upper classes of Lima and other cities enjoyed comfortable lives, the situation of the miners and peasants in the Andean highlands (Sierra) was dire. Land ownership was split between the semi-feudal haciendas that had a home farm and allowed tenants to farm smallholdings, traditionally in return for labour on the home farm; and communities of peasant smallholders, often on the less valuable land. Apart from the mines, the highland economy was stagnant. The difficulties of transport and the limited size of highland cities meant that only small surpluses left the haciendas and smallholdings. Tenancy obligations left many smallholders with little incentive to invest. Moreover, society was split by ethnicity: the land- and mine-owing aristocracy spoke Spanish and dominated economic and political life; while most of the population lived in poverty, spoke Quechua and were subordinated

71

in every way to the elite. In between was a class of mestizo traders and capataces (overseers) who helped the elites maintain this order in exchange for a small share of the spoils.

Given such social inequity, Peru was subject to political strife. The subordinated peasantry found it hard to express their voice, other than through occasional violent revolts. Meanwhile in the mines and cities the working class were unionised and politically inclined to socialist parties. The typical response of the traditional elites was to rule autocratically, making sure they had the support of the military – officers were mainly recruited from the elite – to break strikes or put down rebellions.

For most of the rural population, incomes were pitifully low, health poor and education rudimentary. Most rural women and girls lived hard lives, triply burdened by class, ethnicity and gender.

4.3.2 Development and transformation

Economic development Despite limited and distorted development, by 1960 Peru was, in contrast to most countries in Africa and Asia, already significantly urban and industrialised. Import-substituting industrialisation succeeded in creating an industrial base, but it had at least two flaws. One, national industry tended to be uncompetitive – not least because the domestic market was small, making economies of scale hard to achieve, so few of its products could be exported. Two, the terms of trade were turned against agriculture, from which the majority of the population earned a living. Industrial protection tended to mean costly inputs and consumer goods for farmers, while the exchange rate tended to be overvalued, reducing the value of farm exports – sugar and cotton in particular – and cut incentives for farmers to invest and innovate.

The model was not just economically flawed. It also did little to change Peru’s profound inequalities. On the coast were Lima and other major cities, where much of the manufacturing industry was based. The coastal valleys were mainly occupied by irrigated estates that produced export crops competitively.

Economic weakness and political strife explain the varying performance of the economy from 1960 to the early 1990s (Figure 33). Years of low growth in the 1960s and 1970s corresponded to years of political upheaval, including the 1968 coup that brought Velasco’s left-leaning military regime to power.

The 1980s were marked by the international debt crisis – Peru had racked up debt on the world markets; the ill-fated regime of Alan García who defied IMF recommendations to disastrous effect – and the Sendero Luminoso (Shining Path) uprising in the Sierra that led to civil war in much of the Sierra Sur.

The overall result was that the GDP per person of US$2,600 in 1960 was back at the same figure in 1992, having reached US$3,800 in 1981 and 1987.

The Fujimori regime of 1990–2000 brought in structural adjustment and economic liberalisation. It also helped that the civil war was ended in the early 1990s with the decisive defeat of Sendero Luminoso.

Growth resumed after 1990: half the period has seen annual growth of 5% or more, with only two years of small negative growth. Growth has come partly owing to the high prices for minerals in the 2000s, and partly from flourishing export agriculture on the coastal plain, making Peru a world leader in exporting asparagus and other fruits and vegetables. GDP per person has more than doubled since the early 1990s to reach almost US$6,000 by 2015.

In 1960 the Peruvian economy already was only 20% agricultural, with 30% industry. Since then, the main change has been the declining share of agriculture in the economy, in recent years to less than 10% of GDP (Figure 34).

Demographic transitionPopulation growth has slowed markedly since the 2.5% a year or more seen from 1960 to 1980, to 1.25% since the mid-2000s.

Rural population growth has slowed so much that by the late 1980s the rural areas were seeing absolute declines in population: from a peak of just over 7 million in 1997, currently around 6.7 million live in rural areas. Since rural fertility is higher than urban, out-migration accounts for the declining rural population. The rural share of

72

Figure 33 Economic growth, Peru, 1960–2015

Note: ‘Golpe (de estado)’ is the Spanish term for a coup d’étatSource: Compiled from World Development Indicators

-15

-10

-5

0%

%

5

10

15

GDP annual growth

Golpe Golpe Sendero Garcia

Economic growth

20

50

80

110

140

170

200

GDP at market prices

US$ billions (constant, 2010)

2014

2012

2010

2008

2006

2004

2002

2000

1998

1996

1994

1992

1990

1988

1986

1984

1982

1980

1978

1976

1974

1972

1970

1968

1966

1964

1962

1960

GDP per person

2014

2012

2010

2008

2006

2004

2002

2000

1998

1996

1994

1992

1990

1988

1986

1984

1982

1980

1978

1976

1974

1972

1970

1968

1966

1964

1962

1960

-15

-10

-5

0

5

10

15

0

1

2

3

4

5

6

US$ thousands per capita(constant, 2010)

Figure 34 Structural change, Peru, 1960–2015

Note: No data available for the 1980sSource: Compiled from World Development Indicators

0%

10%

20%

30%

40%

50%

60%

80%

1960

1962

1964

1966

1968

1970

1972

1974

1976

1978

1980

1982

1984

1988

1986

1988

1990

1992

1994

1996

1998

2000

2002

2004

2008

2010

2012

2014

Shar

e of

GDP

Agriculture Industry Services

73

population has fallen from just 50% in 1960 to around 20% in recent years.28

The fertility rate for rural women has fallen sharply since the mid-1980s, from almost 6.5 to less than 3.5, bringing rural fertility much closer to urban levels (Figure 35).

The average age of marriage has risen slightly in rural areas: the average is currently 19.5 years, several years younger than the urban average. The main contributor to reduced fertility, however, is the use of contraception. In the mid-1980s around 25% of married rural women used some form of contraception: that rose steeply, so that today it is 75% of rural women – a rate that matches use by urban women (Figure 36). Correspondingly, the proportion of rural women reporting an unmet need for contraception fell from 36% in 1992 to 10% in 2014.

A major family planning programme was implemented in the mid-1990s, focused on low-income rural women. Reports suggest that health workers used bribes, coercion, and force to meet the high female sterilisation quotas they were given. It is estimated that their forced sterilisation led to women having 0.33 fewer children by 2000, and 0.85 fewer children by 2004, compared to women not subject to the programme (Byker and Gutiérrez, 2012).

Simple inspection of fertility changes in rural areas of Peru and the neighbouring countries of Bolivia and Paraguay shows that fertility decline accelerated in Peru during the programme, but not greatly. Bolivia and Paraguay experienced similar rates of decline without such drastic measures.

Agricultural developmentAgriculture grew only slowly from the early 1960s to 1990, at an annual average of under 2%. Growth was stymied by trade policy that massively discriminated against agriculture (Hermann et al., 1989), and in the highlands, the stultifying effect of the land tenure system in the highlands – see Box 6 and costly transport arising from poor roads. Since 1990 agricultural growth has risen to 5% a year – making Peru’s agriculture one of the top ten fastest-growing worldwide.

28 Peru has a notably low threshold for ‘urban’: any settlement with more than 100 dwellings qualifies. Hence the degree of urbanisation reported is higher than it would be in most other countries.

The fast growth of agriculture since 1990 can be seen in the rapid expansion of fruit and vegetables – most notably export crops such as asparagus, oil palm, coffee, cocoa and tea. For livestock, production has increased since 1990, particularly dairy, eggs and chicken. Five times as much chicken

Figure 35 Population growth, Peru, 1960–2014

0

10

20

30

40

50

60

70

5

10

15

20

25

30

35

40

1960

1964

1968

1972

1976

1980

1984

1988

1992

1996

2000

2004

2008

2012

%

Mill

ions

Rural Total Rural share

Figure 36 Fertility rate, Peru, 1986–2014

Source: Demographic and Health Surveys

2

3

3Fert

ility

rate 5

6

7

1986

1988

1990

1992

1994

1996

1998

2000

2002

2004

2006

2008

2010

2012

2014

Rural Urban

74

was produced by the mid-2010s as in 1990 as chicken has become part of the daily diet.

Most of the rapid increases in production come from coastal, irrigated farms; or in the case of the tree and bush crops, from the eastern slopes of the Andes that lead down to the Amazonian Selva (rainforest). Few products from the Sierra have boomed since 1990. Quinoa, grown largely for export, is a notable exception

as an almost-forgotten indigenous crop has found worldwide appreciation.

The area under crops has more than doubled since 1960, most of the additional land coming from coastal irrigation schemes, and the opening up of land on the margins of Amazonia. The farm labour force – statistics are only available from the mid-1990s – grew until the late 2000s, since when it fell from a high of 3.6 million workers to fewer than 3 million workers in 2014 (Figure 39).

Land productivity was stagnant through to 1990, since when the average value of production has doubled from around US$600 to US$ 1,200 per hectare.

Labour productivity rose in the 1990s and 2000s: almost doubling from an average of US$1,350 per worker in 1997 to US$2,400 in 2012. Average farm earnings are thus probably – one would have to deduct other costs of production to see returns to labour alone – just about enough to escape (US$2 a day) poverty.

Since the mid-1990s, exports of fruit and vegetables have exploded: tonnage up by ten times from 1994/96 to 2011/13, total value up by eleven times. Some 80% of these exports come from, in order of value, grapes, asparagus, avocados, mangoes, mangosteens, guavas, bananas, and citrus fruits (tangerines, mandarins, clementines, and satsumas).

Most agricultural growth since 1990 can be attributed to private initiative and investment. The state has made this possible by ensuring macroeconomic stability and a liberal economy that encourages investment. The exceptions concern roads and irrigation. The opening up of the Amazonian margins was made possible by improved roads that connect the fringes of the Selva to the coast. On the coastal plain, expanded irrigation has contributed to the export boom: since 1990 another 1 million hectares have been irrigated to add to the existing 1.6 million hectares. Much of the additional irrigated land comes from large public schemes.

Peru undertook radical land reform from 1969 to 1975 redistributing land from landlords to farm workers, tenant farmers and smallholders. It released many peasants from the grip of the landlords in the Sierra, but otherwise did little to increase production (Box 6).

Figure 37 Use of contraception by adult married women aged 15–49, Peru, 1986

Source: Demographic and Health Surveys

20

30

40

50

60

70

80

1986

1988

1990

1992

1994

1996

1998

2000

2002

2004

2006

2008

2010

2012

2014

%

Rural Urban

Figure 38 Agricultural growth, Peru, 1961–2013

Source: Compiled from FAOSTAT data. Three-year moving averages taken to reduce the effect of yearly variations in weather

1

2

3

4

5

6

7

8

1962

1968

1974

1980

1986

1992

1998

2004

2010

Gros

s va

lue

US$

billo

ns

(con

stan

t, 20

04/0

6), 3

YMA

75

Rural transition, labour and rural wagesEconomic growth in the 1990s and early 2000s did little to improve the lives of those on low incomes. Not only did poverty rise, but also inequality widened (Figure 40). The jobs that were created at the time seem to have been largely informal and poorly paid. Indeed, average wages paid by private firms in Lima in 2000 have been estimated to be just 20% of those paid in 1974 in real terms (Plaza and Stromquist, 2006).

Subsequently, there have been some quite remarkable reductions in poverty. Rural poverty fell from more than 80% in 2005 to 30% by 2015. Examining these changes for 2004 to 2010, that is before the full extent of poverty reduction was seen, Inchauste et al. (2012)

found that most of it be could be attributed to a combination of people on low incomes spending less time on agriculture and more on non-farm activity, and on rising productivity and returns to labour rather than increased skills.

Additional jobs were created on a large scale from 2000 onwards. The coastal estates generated another 60,000 jobs for unskilled workers between 2000 and 2010. Given an agricultural work force estimated at between 2.5 and 4 million, that is a sizeable increase. Work on the estates is mainly casual and seasonal: half the workers are migrants – mainly from the Sierra, and half are predominantly young women of an average age of 20 years. Labour shortages have been reported at peak times: reported wages

Figure 39 Agricultural area, labor, and productivity, Peru, 1961–2013

Source: Compiled from FAOSTAT data

1,500

2,000

2,500

3,000

3,500

4,000

4,500

5,000

5,500

6,000

500

1000

1500

2000

2500

3000

3500

4000 Area, Arable & Permanent Crops, k haAg

ricul

tura

l La

bour

For

ce,

k pe

rson

s

Men Women Total Arable & Permanent Crops

500 700 900

1,100 1,300 1,500 1,700 1,900 2,100 2,300 2,500

Valu

e Pr

oduc

tion,

US$

con

stan

t 20

04/0

6,

per

pers

on o

r he

ctar

e

Labour productivity Land productivity

76

are almost 20% above the legal minimum wage (Schuster and Maertens, 2015).

In addition, since 2004, the government began to spend mineral royalties on education, health, water, social transfers, roads and irrigation across the country, but especially in the Sierra, making significant inroads to correct the centuries-long

deficit the highlands had experienced. Increased spending in rural areas has been large in terms of the fraction of GDP invested (Figure 41). Increases in absolute terms are even sharper: between 2002 and 2012, spending per person in rural areas rose by almost six times (Fort and Paredes, 2014). Moreover, returns on investment

Box 6 Peru’s radical land reform: changing structures but not productivity

Following a coup in 1968, a left-leaning military government came to power intent on reforming land tenure for reasons both of national sovereignty and social justice. At the time, 75% of the land was held by estates of 500 or more hectares, just 1% of holdings; while 82% of farms had less than five hectares and occupied just 16% of the land. On the coast, some of the estates were held by foreign companies.

In 1969, the government passed an agrarian reform law, designed to break the power of the landlords, especially of the foreign owners of the sugar estates; provide more land to peasants; and to stimulate investment and production. Properties of over 150–200 ha on the coast and 15–330 ha in the Sierra could be expropriated by the state, potentially affecting 10M of 24M ha of agricultural land. Compensation was mainly by bonds with 20 years’ maturity, bearing 5% interest, not inflation-indexed, based on land values determined by the declared value in the 1968 tax returns.

The coastal sugar estates were promptly taken over by the military. Nominally cooperatives belonged to the former workers, but the government appointed managers and technical staff and continued to operate them as estates: the members were entitled to dividends. It was not long before the model disintegrated owing to a combination of poor management, high operating costs, and the state trying to recuperate debts and taxes from the estates. By the late 1980s the state gave up: the cooperatives were split up into individual holdings.

In the Sierra, peasants quickly seized some estates, but otherwise the reform progressed slowly. Given that the large estates included plenty of marginal pastures, there was far less arable land to redistribute than had been imagined. The government tried to set up peasant groups and other cooperative forms of cultivation, but as with the coastal estates these groupings struggled with debt, unfavorable terms of trade, and collective management. In reality, the peasants wanted their own plots, not cooperative farms. With time, the cooperatives were dismembered as peasants took their own land.

By mid-1979, when the military had once again given way to a civilian administration, 40% of the land had been given to 35% of farming families. Despite the scale of reform, it achieved only the political objectives of breaking the landlords’ grip on rural society. Investment and production languished in view of the unfavorable economic conditions, the lack of complementary public investments in roads and irrigation, and supply of inputs, and collective farms proved hard to manage.

The reforms were officially ended by the 1995 Ley de Tierras that removed ceilings on land ownership and restricted the scope for expropriation; as well as making state-owned lands accessible only through public auction. The subsequent individualisation of tenure meant that by the 1990s a titling program began, the Proyecto Especial de Titulación de Tierras y Cadastro Rural – PETT.

Sources: Kay (1982), Larson et al. (2000), Mayer (2009), Melmed Sanjak et al. (1991), Mosley (1985), Zaldívar (1974)

77

in social support, roads, education, producer support and irrigation have been good (ibid.).

4.3.3 Changes to the lives of rural women and girls

Women as economic participantsSince 1990, rural women have had more opportunities to earn an income. One avenue has been the growth of the rural non-farm economy (RNFE) as improved roads have encouraged the growth of towns and small rural centers. Some of the jobs on offer, such as trading in markets, provide as many if not more opportunities to women than to men.

Another route has been migration. A generation or two ago, most migration from areas was by men, with wives and mothers left behind to run the farm and household. The 1980s saw mass rural displacement in the conflict-affected central and southern Sierra: as many 600,000 moved to Lima and other urban centers (Barrios, 2014). Not all the displaced returned when peace was restored. The links between migrants and their family and community have subsequently facilitated voluntary migration. Net out-migration may be less than during the 1980s conflict, but remains high from the poorer parts of the Sierra. For

example, between 2002 and 2007 more than 5% of the population of the highland departments of Apurimac, Cajamarca, Huánuco, and Huancavelica migrated (Yamada, 2010). Around half of the migrants were women and girls.

As women increasingly entered the labour market, they sought better opportunities, just like the men. Women also found good opportunities in the city both in commerce and as domestic workers. In the countryside, women found increasing educational opportunities and changing gender roles prepared them to enter the labour market. Thus, migration allowed women to discover better horizons and even better marriage prospects. [Altamirano, 2003]

The agricultural censuses show that the number of farms run by women almost doubled from 354,000 in 1994 to 669,000 in 2012.

Figure 40 Poverty and inequality, Peru, 1985–2015

Source: Poverty and Equity Database, World Bank. Poverty measured by national poverty lines

20

30

40

50

60

70

80

90

Perc

enta

ge p

oor/

Gin

i sco

re a

s %

Gini coeff. Rural poverty National poverty

Figure 41 Growth in public investment in rural areas as a percentage of GDP, 2002–2012

Note: Infraestructura productiva = Production infrastructure; Infraestructura social = Social infrastructure; Apoyo directo a productores = Direct producer support; Apoyo social = Social support Source: Fort and Paredes (2014), based on data from the Integrated Financial Administration System (SIAF) of the Ministry of Economic Affairs and Finance (MEF), the National Public Investment System (SNIP) and the Central Reserve Bank of Peru.

78

This seems to have arisen as men have left the villages, leaving more women in control of the farm (Espinoza, 2016).

Women as mothers, wives and carersSeveral factors have reduced the burden of domestic work and child care on rural women.

One is the fall in fertility, as reported above: rural rates have almost halved in the last 30 years. Moreover, women may well be spending less time caring for sick infants as under-five mortality rates have fallen by 80% since the mid-1980s (see next section).

Improvements in housing have reduced some of the domestic work. By 2014, 74% of rural homes had electricity and 62.5% had running water: in 1991/92 the respective statistics were 20% and 18%.

Since 2012, Peru has had a day-care programme for children under three from low-income rural families , called ‘Cuna Más’. In 2012 it was attended by some 70,000 children (Lavigne, 2013).

Women’s human developmentIn the 1980s, rural Peru had very high indicators of ill health. Those have improved, although rarely to the levels one might associate with an upper-middle-income country. For example, the national maternal mortality ratio – no rural estimate has been published – fell from 239 per 100,000 live births in 1992 to 157 in 2011. The upper-middle-income average in 2014 was 57: neighbouring Ecuador had a rate of 70 in 2014.

There has been greater improvement in the under-five mortality rate, which in rural areas fell from 160 per 1,000 live births in 1986 to 32 in 2014, approaching urban levels (Figure 42). Stunting among under-fives in rural areas also fell from 55% in 1992 to 29% in 2014, although this remains high for a country with Peru’s resources.

Rural women’s education has increased. Their literacy rates reached 94% in 2014, only a little less than in urban areas: in 2000 the rate recorded was 78% (Figure 43). The share of rural women who have secondary or higher education has reached almost 50%, up from 27% in 1992, but remains considerably less than in urban areas. This share will continue to rise since rural girls’ net attendance at secondary

school had risen to 62% by 2012, only a couple of percentage points less than that for boys, or for urban areas.

Rural girls’ secondary schooling corresponds as much to rising demand as supply: rural parents and their daughters are coming to see the value of girls’ education (Box 7).

Women as citizens and agents, gender normsWomen in Peru have for long had secondary status compared to men, their lower position in social hierarchies as women intersecting with other marked differences by class, race and language. One manifestation of this has been high rates of sexual violence, apparently higher than averages for Latin America, directed mainly at women and minors (Mujica, 2011). The conflict of the 1980s and 1990s in the highlands saw rural women particularly affected, with much sexual violence used by combatants against rural women (Boesten, 2012).

The state has formally tried to improve the status of women in law, to reduce violence, and provide services, not least under the Fujimori government of the 1990s.

… Fujimori started to use a highly gendered political strategy, speaking to feminist demands for greater women’s participation in politics, a ministry for

Figure 42 Under-five mortality, Peru, 1986–2014

Source: Compiled from Demographic and Health Surveys

10

30

50

70

90

110

130

150

170

Deat

hs p

er 1

000

befo

re A

ge 5

Rural Urban

79

women’s affairs, better reproductive healthcare, and access to birth control (culminating in a forced sterilization programme), a (clientelist) food dis¬tribution programme directed at Peru’s “self-sacrificing mothers,” and indeed, interventions to tackle violence against women. (Boesten, 2012)

It did so, however, within the framework of authoritarian politics in which it was assumed that men would control the lives of women. The result has been that implementation of measures to protect women has been weak.

… the system that is to protect women from violence is largely staffed with people who prefer to identify with an existing ideology in which men dominate women, and women are expected to endure certain levels of violence, rather than being ‘feminist’. (Boesten, 2012)

In particular, the approach adopted by the women’s ministry has been to privilege the family over women’s rights, and therefore, in as much as most households are headed and controlled by men, to reinforce patriarchy. (Boesten, 2012)

Despite this unpromising background, some indicators suggest that rural women’s status may be improving in some respects, albeit from a low base. For example, the most recent DHS of 2014 shows a drop in rural women reporting spousal violence, from 38% in 2004 to 31% in 2014. It is an improvement, but still leaves almost one third of rural women exposed to male domestic violence.

There also signs are that women are more often getting leadership posts, albeit again from a very low starting point. A 1997 law stipulates that all lists of candidates for elections at all levels must contain at least 30% of each sex. At municipal level, the share of women councilors has risen from 6% in 1983 to 43% in 2015. But no quotas for government leadership have been set, so the share of women has barely changed, at 2–4% of the total (Espinoza 2016).

Land and property laws are clear: for those dying intestate, 50% goes to the surviving spouse and 50% to their offspring, divided equally. In practice, the rural custom is to allocate more to sons than to daughters. When the PETT land titling programme (see Box 6) was introduced, couples were meant to register both names. Yet from 1992 to 2005, teams would often cut corners by registering just the man, until there was a public outcry to rectify this. It did not

Figure 43 Rural women’s education, Peru, 1992–2014

Source: Compiled from Demographic and Health Surveys

75

80

85

90

95

100

Rural Urban

75

80

85

90

95

100Rural Urban

80

help that many rural women lacked a national identity card (DNI) when it came to registering their rights (Espinoza, 2015).

4.3.4 Conclusion: labour markets and public servicesPeru’s development shares some characteristics with Thailand. The lives of rural women and girls have improved through two main channels:

Box 7 Higher aspirations for rural girls

In the past, girls who attended rural secondary schools were often subjected to denigration and prejudice because of their sex, class and language, to the point where they just left. Becoming a young wife and mother in their village conferred some status and respect.

Recent reports from studies that track the lives of young Peruvians indicate that this is changing. Some 670 children have been tracked in the older cohorts of the ‘Young Lives’ study.

Parents who might previously have regarded it as unnecessary to send their daughters to school now see that opportunities in farming are limited and that there are many non-farm jobs, which are often better paid. Thanks to expansion of public services in rural areas, they regularly meet women workers in health, education and other services. This has raised their aspirations for their daughters, but they know that the only way to obtain such jobs is through formal education. The girls may still suffer discrimination at school, but they and their parents know that they have to ignore this and get on with their studies (Ames, 2013).

If other factors such as poverty, ill health and low school quality continued to menace rural girls’ education, there was one important thing that had changed in significant ways: the attitude towards girls’ education, among their families, but also among girls themselves (Ames, 2013).

A turning point may well have been reached: one that promises much for the future, not just for education, but also for identity and self-respect:

Current interest in the construction and transformation of women’s identities suggests that far from being superficial, the change of attitudes emerging among rural girls in this study may reflect turning points in the ongoing construction of their very identities. Indeed, the ‘educational desire’ of young women from marginalised groups has been one major driving force for change and transformation in women’s lives, as Mirza (2006) reminds us, and a way to contest and challenge the stereotypes about them, (Ames, 2013).

There are cultural contradictions involved: it is not just a matter of becoming educated. For a young rural woman to become somebody she must abandon her indigenous clothes, language and identity. In addition, most good jobs are in the towns, which entails migration.

Thus, to overcome the various inequalities that affect their lives (persistent poverty, gender oppression, and ethnic discrimination) rural and indigenous girls pursue an educational career that could entail stopping being rural and looking indigenous, and postponing or abandoning marriage and motherhood (Ames, 2013).

Realising their new goals is not easy: they need money and good health, while avoiding the temptations of bad company or substance abuse, or becoming pregnant.

Source: Ames (2013)

81

one being economic growth, structural change, and increasing opportunities for women to work or run businesses; the other arising from much improved rural services.

In these respects, Peru has seen similar changes to those seen in Thailand, but to a lesser degree. Inequality is deeper in Peru: the neglect of the rural Sierra has historically been so great, that even with the intensified efforts of the last ten or more years, rural living conditions remain far inferior to those of the cities on the coast.

As in Thailand, although the material conditions of rural women’s lives have improved, changes in gender norms and status have been slower.

For this study, interviews were carried out with seven experienced Peruvian observers of rural areas, some of whom previously managed programmes, one a former minister, four of them women. They were adamant that rural women’s lives were improving, although some stressed that even those programmes that benefited women most directly – particularly the Juntos conditional cash transfer (CTT) programme29 – were not designed to achieve gender equality. The intent was primarily to reduce poverty and encourage use of schools and health clinics.

29 In May 2015 Wiggins met those implementing this programmeme at national, regional and field levels. Almost all the staff met, from senior policy makers to field workers, were women.

This insight can be read in two ways. It may indicate how far Peru has to go before gender equality is built into public programmes as an accepted and primary objective. It may however also indicate how much can be done to support rural women, without necessarily having to the win the argument for gender equality.

The observers saw the following measures as the next steps for improving the lives of rural women and girls:

• Ensure that at school children receive sex education, that women’s roles and rights are taught, and that role models of women as professionals, scientists are projected.

• Ensure that girls can attend school securely – especially important for boarding pupils.

• Ensure women’s rights to land are respected, as legally required, through specially-trained agrarian magistrates.

• Create more quotas for female leadership. • Train public officials on gender perspectives and rights.

4.4 Egypt: intractable conundrums?

4.4.1 BackgroundIn the early 1950s Egypt was largely agrarian, with most of the population working in agriculture. Much of the land belonged to landlords, most farmers were tenants and landlessness was common. Out of 2.8 million landowners, 94% of them possessed less than five feddan (2.1 ha) making up just 35% of the land area. The economy was growing slowly, and the land-tenure system allowed landowners to rack up rents (Bush 2005). Poverty was rife and endemic.

Although Egypt had formally recovered independence in 1936, British troops remained for another 20 years. Effective independence can be variously dated to 1952, when the Free officers overthrew King Farouk, 1954 when Nasser became president, and 1956 when the British military presence ended and the Suez Canal came under Egyptian control.

Figure 44 Reported spousal violence, Peru, 2004–2014

Source: Compiled from Demographic and Health Surveys

30

32

34

36

38

40

42

44Rural Urban

82

Nasser followed a state-led, Arab socialist growth model with support from the then USSR. Large industries and banks were nationalised. Public service jobs were guaranteed for all graduates. Large-scale public works were undertaken, chief among them being the Aswan High Dam (Ghanem, 2015).

Not least of Nasser’s programmes was land reform: enacted within two months of the 1952 coup, backed up by further legislation in 1961, 1963, and 1969, the reforms imposed a land ceiling of 200 (84 ha), and later 100 feddan (42 ha). Land in excess was redistributed to those with smallholdings, agricultural tenants and the landless: some 2 million persons benefited by 1962. Land rents were controlled, tenants had secure and heritable tenure. The government organised cooperatives to ensure that smallholders could get access to inputs, credit and markets.

4.4.2 Development and transformationSince 1960, Egypt’s growth and development has been punctuated by sharp changes in policy and in external circumstances. These comprised:

• war with Israel in 1967 and 1973, a state of hostility that ended with the Camp David accords of 1978

• a realignment of external relations in the early 1970s, as the friendship treaty with the USSR was ended in 1971 and Egypt aligned with the USA

• the economic opening up and liberalisation of 1973 (‘the infitah’)

• the oil price hikes of late 1973 and 1979 • further economic liberalisation in the early 1990s.

Owing to such events, growth has been uneven since the mid-1960s. No less than five spells when domestic product per person fell can be seen: three of these occurred when wars disrupted the economy (1967/68; 1973/74; 1992), the 1987/88 slump was the product of a debt crisis; while output per person fell after the Arab Spring of 2011. These have been

30 Growth since the 1960s has been rapid compared to earlier periods. Between 1870 and 1967, incomes rose 1.8 times, at an average of just 0.6% a year (Maddison World Table). Since then, growth per person has averaged 2.7% a year.

offset by some periods of rapid growth. Most notably in the late 1970s Egypt benefited from high oil prices and the boom in the Gulf States to which several million Egyptian migrant workers moved (Figure 45).

Overall, between 1967 and 2015, GDP per person rose by 3.6 times in constant terms, taking Egypt out of the low-income category.30 In the process, Egypt has become an industrial economy: agriculture’s share of GDP has fallen from 29% to 11%, while industry’s share has risen to 39% (Figure 46).

Egypt’s economy and its growth, however, have depended heavily on volatile sources. Since the oil price rise of 1973, oil has not only generated 40% of export income, but also millions of Egyptians have worked in the Gulf and Libya. Those jobs have not only been vulnerable to fluctuating oil prices, but also jobs have been lost when conflicts have broken out. Aid receipts have been copious: enough to finance 20% of government spending in the 1990s. Tourism has generated as much as 20% of foreign exchange earnings; but this sector has been badly affected by terrorist attacks. Less volatile have been earnings from fees for the use of the Suez Canal (Ghanem, 2015; Salem and Gleason, 2004).

Egypt has not been able to develop export manufacturing in the way that some Asian economies have done, despite its large labour force.

Growth since the liberalisation in the early 1990s was quite strong until 2011. Questions, however, have been raised about how inclusive this growth has been. Privatisation of state industries has often meant informalisation of employment, with few benefits and lower wages – at very least for new recruits. At the same time, the large public budget that underpinned the corporate model that started in the 1950s, has meant reduced subsidies on basic goods, and pressure on public services (Mansour, 2016).

Demographic transitionEgypt’s population has grown quickly: from 27 million in 1961 to more than 90 million by 2015

83

(Figure 47). Growth rates, although fluctuating – in the late 1960s and early 1970s owing to war – have barely slackened from 2.7% a year seen in the early 1960s to more than 2% a year seen in recent years. For a middle-income country, recent population growth has been surprisingly strong.

While death rates have fallen and life expectancy has been increased from 49 to 73 years for women, fertility has fallen only slowly. In 1988 the rate in rural areas was 5.6, which had fallen to 3.8 by 2014. Even urban rates are close to 3 (Figure 48).

It seems that the demographic transition in Egypt will be protracted, meaning that every year a large youth cohort enters the labour market.

The immediate causes can readily be seen. While married women’s use of contraception increased from 25% in 1988 to 57% by 2004, uptake in rural areas – and also in urban areas – has stalled at just under 60% for rural areas (Figure 49).

This is not for lack of official encouragement: a public family planning programme exists. The problem probably lies with the way this programme is delivered: family planning advice and contraception are dispensed at health centers, rather than on the doorstep by paramedics, as applies in Thailand. Bangladesh, which shares some characteristics with Egypt – a dense and majority-Muslim population – has achieved higher uptake of contraception by using doorstep services (Amin and Lloyd, 2002).

If contraception has not been that widely adopted, average age of first marriage in rural areas has increased, from 17 to 20 years between 1988 and 2014.

Urbanisation appears unusually low in Egypt: the share of the population apparently living in urban areas has grown from 38% in the early 1960s to only 43% in 2015.

This, however, probably considerably understates urbanisation. Egypt defines urban as living in a district capital or one of the five large cities. Not only does that omit smaller

Figure 45 Economic growth, Egypt, 1967–2015

Source: Compiled from World Development Indicators

-4

-2

0

2

4

6

8

10

12

14

500

1,000

1,500

2,000

2,500

3,000

US$

per

cap.

, co

nsta

nt 2

010

Growth per cap, % pa GDP per cap. constant 2010 US$

Figure 46 Structural transformation, Egypt, 1966–2015

Source: World Development Indicators

0

10

20

30

40

50

60

Shar

e of

GDP

, %

Agric Industry Services

84

towns –there are ‘villages’ in the Census with more than 100,000 inhabitants; but also the boundaries of the cities have not been revised sufficiently – half of the population of Cairo, for example, lives beyond what is officially the city, in what administratively are rural areas. The reality is that the overwhelming majority of Egyptians live in the Nile Valley and Delta at population densities of more than 1,000 persons per square kilometer, with ready access to large urban centers.

Agricultural development Agricultural output has grown quite strongly since the early 1960s, rising more than five times in total (Figure 50). A clear break can be seen in the rate of agricultural growth. From the early 1960s to 1989, growth averaged just 2.1% a year, barely keeping pace with population growth. From 1990, however, it rose to more than 3.8% a year, considerably ahead of population growth.

The acceleration arose when reforms to agricultural policy started in the early 1980s took effect. Reforms saw the ending of quotas for planting and production of some of the main crops, and of state procurement of cotton, sugar and other crops at prices below world market levels. Subsidised inputs and credit were also largely curtailed, even if fertiliser, made domestically by a parastatal, remains subsidised.

Increases since 1990 have been especially marked for livestock products: production of eggs and chicken in particular has increased by more than dozen times between 1990 and 2013. Increased poultry production has required large volumes of grains as feed.

Both land and labour for agriculture have been considerably increased since the early 1960s (Figure 51), despite the apparent lack of additional land to bring into cultivation. The Nile plain and delta have long been fully cultivated, yet more than 50% has been added to the cultivated area. This has come from a combination of new irrigated areas made possible by the Aswan high dam and some groundwater irrigation; and by reclaiming land in the delta that had become waterlogged and

Figure 47 Population growth, Egypt, 1961–2015

Source: World Development Indicators

1.5

1.7

1.9

2.1

2.3

2.5

2.7

2.9

10

30

50

70

90

110

Annual growth, %

Annual Growth Rate Rural population

Total Population

Figure 48 Fertility, Egypt, 1988–2014

Source: Compiled from DHS

2

2.5

3

3.5

4

4.5

5

5.5

6Rural Urban

Figure 49 Married women using contraception, Egypt, 1988–2014

Source: Compiled from DHS

20

25

30

35

40

45

50

55

60

65

70

Rural Urban

85

saline. Since 1980, another 30% (712k ha) have been added to the irrigated area.

The farm labour force has also risen, from 3.5 million in the early 1960s to 6.5 million by 2013.31

Agriculture has intensified. Land productivity has increased by more than four times since the early 1960s. Egyptian yields are among the highest in the world: first for rice (10 tonnes/ha), tenth for wheat (6.5 t/ha), and twentieth for maize (8 t/ha). Although labour productivity scarcely rose at all until the late 1980s, since then it has more than doubled.

31 The considerable fluctuations in the agricultural work force result mainly from (mainly male) migration out of villages in response to oil-induced booms, and returns when oil prices have fallen or conflict has broken out.

Rural diversification, the rural non-farm economy and migrationIn a country where the ratio of rural population to land is high, with the median farm size being less than 0.5 ha, it is not surprising that for decades rural households have been looking for more and better rewarded off-farm work. Some of that work can be found locally, in the non-farm part of the village and rural economy. In 1997, a national household survey showed that on average, crops and livestock were providing less than 30% of rural household income (Croppenstedt, 2006). Many of the rural

Figure 50 Agricultural output, Egypt, 1962–2013

Note: Chart shows a three-year moving average to reduce the yearly variations in weather.Source: Compiled from FAOSTAT data.

3.5

5.5

7.5

9.5

11.5

13.5

15.5

17.5

Gros

s Va

lue

Prod

n, c

onst

ant

US$

2010

, Bi

llion

2.1% a year

3.88% a year

1.01.52.02.53.03.54.04.55.05.56.0

-

200

400

600

800

1,000

1,200

Milk, m

illion tonnes, 3YMA

Mea

t or

Eggs

, x

1,00

0 to

nnes

, 3Y

MA

Eggs Sheep & Goat Meat Beef Poultry Meat Milk

86

non-farm jobs, however, depend on linkages from agriculture (Mellor and Gavian, 1999).

It is not only local jobs: migration has long been an important part of the lives and livelihoods of many rural households. Historically, rural work gangs recruited from villages in the off-season provided labour for roads and canals, and for building mega-projects such as the Aswan High Dam.

After 1973 and the oil price hike, millions of Egyptians migrated to neighbouring countries, above all Saudi Arabia, Libya, Iraq, Kuwait and Jordan. Although many of these migrants were from urban Egypt, since they had the skills required in the destination countries, their departure opened up jobs for rural labourers to move to urban building sites – construction funded in large part by migrants’ remittances.

Emigration from Egypt has varied: it peaked at more than 3 million in the early 1980s, then fell with declining real prices for oil and the first Gulf War, although by the mid-1990s more than 2 million Egyptians were estimated to be working abroad (Zohry, 2005).

For many rural households, migration – almost exclusively male – has become critical for livelihoods. Men move seasonally and temporarily to the cities looking for work in construction and other informal activities. Often, they may be poorly paid, but the rewards are

better than they can earn in the village. This applies especially for Upper Egypt, where land is particularly scarce and there is heavy competition for off-farm work. Toth (2002) paints a vivid picture of the way that this migration has swamped the urban job markets:

[Rural migrants after 1973 inundated Cairo, Alexandria, the Canal Zone cities, and provincial capitals and towns to find jobs that were, simply by definition, forever better than those back in the village. Replacing workers who emigrated abroad in response to the oil boom and then later emigrating themselves, they emptied the Egyptian countryside (relatively speaking) and inadvertently created an urban surplus labour army that wreaked havoc with Egypt’s labour markets. Urban Egypt was still unprepared for such substantial employment, and so workers squirmed and squeezed their way into informal sector activities, involuted in the ways that Clifford Geertz once depicted for Indonesian agriculture (1963), but now located

Figure 51 Agricultural area, labor and productivity, Egypt, 1961–2013

Source: Compiled from FAOSTAT data

3

3.5

4

4.5

5

5.5

6

6.5

7

7.5

2.2

2.4

2.6

2.8

3

3.2

3.4

3.6

3.8

Million personsm

illlio

n ha

Agricultural Area Agricultural Labour

500

1,000

1,500

2,000

2,500

3,000

3,500

4,000

4,500

5,000

Land Prody Labour Prody

87

in urban interstices. (Toth, 2002: 4, paragraphing added)

‘Upper Egyptian labourers’, comments Zohry (2005),

live a miserable life in Cairo in order to ensure a decent life for their families. … This type of migration is totally male-dominated. It is not socially acceptable for women to undertake such migration.

Labour markets, employment and povertyCentral to discussions of Egypt’s development is employment, a longstanding concern that has grown since the 1990s. In the 2010s, many of the country’s youth have limited prospects of finding a job, with long delays between leaving education and finding employment. Even then, many jobs are informal and badly paid. Above all, the prospects for young women are considerably worse than those for young men (Assaad and Barsoum, 2008; Ghanem, 2015).

This dire situation has roots in both supply and demand. Rapid population growth means that every year a large youth cohort enters the job market: between 750,000 and 850,000.32 On demand, economic growth that picked up from the mid-1990s has not generated a commensurate number of jobs. Between 1997 and 2005, the economy grew by 4.2% a year on average, yet jobs were created at the rate of just 2.5% a year – below the growth of labour supply at around 2.9% a year (Hassan and Sassanpour, 2008).

Lack of jobs stems from a combination of even fewer jobs being generated in manufacturing as

32 In 2016, Egypt has almost 9 million aged 15 to 19, suggesting that as many as 1.8 million more youth are potentially added to the labour supply every year. Just 2.4 million are aged 60 to 64 years, suggesting that no more than 0.5 million a year likely to retire.

The current cohort of 0 to 5 year olds is 12.8 millions, which suggests that if anything, additions to the labour supply may increase in the near future.

33 This is especially disappointing because the state has created social protection to prevent this, but seemingly to little effect:

‘The Egyptian government has not developed a national strategy to reduce poverty. As a result of the legacy of earlier populist policies, the government does possess a sophisticated social protection system that is highly inefficient and transfers most benefits to the better-off rather than to the poor. Some programmes, notably bread subsidies, have had a positive impact on the poor through the introduction of self-targeting to reduce leakage’ (Salem and Gleason, 2003).

the sector grows, while there has been insufficient growth in those sectors that tend to generate most jobs: agriculture, construction and services (Hassan and Sassanpour, 2008).

A contributory factor may be that agricultural growth since the 1990s has not generated the anticipated jobs, because the crops and livestock that have expanded since then have tended to be produced by medium or large-scale holdings, rather than smallholdings. These may also have fewer links to the rural non-farm economy.

Consequently, since the 1990s poverty has, if anything, increased. The second half of the 1990s saw some small reductions in poverty, but these were not evenly spread. Improvements for rural households were marginal (El-laithy et al., 2003). Since then, rural poverty has risen from 22% in 2000 to 32% in the most recent estimate, 2010 (World Bank Poverty and Equity database).33

4.4.3 Changes to the lives of rural women and girls

Women as economic participantsRural women in Egypt are almost entirely restricted to work in the farm household – as much as 85% of all women’s work (Chen et al. 2005; Joseph and Slyomovics 2011). Some rural women may also work on other local farms, and some may trade in local centers. Although working in the fields does not accord with some idealised norms, nevertheless it seems that women have always done so, except for farm operations seen to demand male strength. Women are particularly engaged in livestock, fruit and vegetable production (Abdel Aal, 2002).

When women farm, they have less access than men to inputs, credit and technical advice (Abdel

88

Aal, 2002). From the late 1950s the government organised farmers into cooperatives to facilitate state support and direction of small-scale farming; but most cooperative members were men.

Migration from villages, almost entirely by men, has reinforced this: small farms in Egypt depend heavily on women’s labour and management (Toth 1991; Binzel and Assaad 2010). Women may also undertake wage labour on local farms: they may be preferred to male labour since it is possible to pay them up to 50% less (Toth 1991).

Traveling from al-Minya, Egypt, north to Cairo by train in the mid-1980s, an observer could not help but notice the large numbers of women and children working in the lush agricultural plots adjacent to the railroad tracks. In fact, throughout the northern Sa’id and the Delta, too, women seemed to have taken over the fields. Furthermore, upon arriving at the Agriculture Ministry in Giza, a listener frequently overheard remarks referring to the “alarming” rise in the feminisation of Egyptian agriculture (Toth, 1991).

In the past, some rural girls were able to escape these limitations through education to secondary level or beyond, at which point they may have been able to get a formal job. The public service and state-owned factories had quotas for women employees. Liberalisation since the 1990s, which has led to the Privatisation of some state firms, and attempts to limit public services, has made these jobs scarcer.

Moreover, Egypt’s liberalizing economy has taken some unusual turns. New jobs in assembly plants, textile and garment factories and the like, so often seen in emerging economies and which are often taken up by women, have not been created on any significant scale. Instead:

It appears that in Egypt, employment growth was more rapid in disproportionately male sectors, such as transportation, construction, utilities, and a variety of service industries, which are typically considered

non-tradable. Moreover, there was no tendency toward the feminisation of these male-dominated sectors in Egypt, while the disproportionately female sectors other than the civil service de-feminized (Assaad, 2004).

Even in sectors that do generate jobs for women, such as tourism, Egyptian hotels and restaurants are staffed almost exclusively by men. The problem is largely that of movement: it is simply not possible for most rural women to commute to jobs, let alone migrate (Assaad, 2002; Assaad and Arntz, 2005)

Barriers to entry to waged employment mean that women are crowded into a few market segments, which depresses wages (Assaad and Arntz, 2005).

The participation of women in waged work is very low in Egypt:

Compared to 63 percent of young men, 22 percent of young women fifteen to twenty-nine were economically active in 2006. This gender gap in labour participation is among the highest in the region, if not the world (Assaad and Barsoum, 2008).

The likelihood of rural women obtaining a job a job is slim for those who did not complete primary school. Geography counts even more: rural women in Upper Egypt participate in the labour force to half the extent of rural women in Lower Egypt (Assaad et al., 2000).

Women as mothers, wives and carersWomen’s lives as carers may have improved a little. Rural fertility has fallen, even if only from 5.6 to 3.8; so most rural women should have fewer children to care for. Probably more important are significant improvements in the health of infants and young children, so women may be spending less time caring for sick children.

Housing conditions have improved. In 1992, 45% of rural households had running water and 83% had electricity; by 2014, those proportions were 90% and 99%. In 1992, Upper Egypt has significantly lower incidence for both services: by 2014 almost no difference could be seen between

89

rural areas of Upper and Lower Egypt ([DHS reports 1991 and 2014).

Despite some reduction in the work that almost always falls to rural women and girls, the combined burden of productive and reproductive work can be heavy in rural Egypt, admittedly looking at the most deprived part, Upper Egypt:

[In Upper Egypt] Rural women may spend up to 19 hours a day performing essential chores such as sowing, weeding, harvesting, animal husbandry, fetching water and firewood, and other home activities. Women’s access to and control over productive resources (inputs and labour-saving technology) and support services (credit, extension, training and markets) is limited (IFAD, 2006).

Women’s human developmentThe health of rural women and girls has improved considerably since the late 1980s: taking the deaths of children before their fifth birthday, the rural rate has fallen by almost 80% since 1988 (Figure 52).

For the country as whole, the maternal mortality rate has been reduced from more than 100 deaths per 100,000 live births to 33 between 1988 and 2015. DHS surveys indicate why: in 1988 in rural areas, just 12% of women received regular antenatal care and only 19% delivered their babies in medical facilities: by 2014 those proportions were 80% and 89% respectively.

Education for girls has improved as well. By 2014, almost two-thirds of rural women were literate; and almost 60% of rural women had been to secondary school or higher (Figure 53). Expansion of schooling is taking effect and gap between girls and boys going to school has narrowed.

Two important qualifications apply, however. One is that glaring differences show up by wealth and region. For example, girls’ enrolment in secondary schools from the richest quintile rose from 52% to 72% between 1988 and 2005, with hardly any difference between girls and boys. For the poorest quintile, however, the increase was from 1.5% to just 6.6%, at half the level seen for boys from this income group (Cupito

and Langsten, 2011). Not surprisingly, given low incomes in Upper Egypt, this region also records significantly lower rates of girls’ educational enrolment and attainment.

It seems that for children who are disadvantaged by coming from a low-income household, by living in rural areas and in Upper Egypt, and by having parents who lack education, have very poor chances of entering and completing secondary education. Moreover, this does not seem to have improved much in the last 20 years: indeed some fear the differences may be widening (Akkari, 2004; Ersado, 2012).

The other qualification concerns the low quality of much public education, particularly in rural areas. Schools do not produce marketable

Figure 52 Mortality of children under five years, Egypt, 1988–2014

Figure 53 Rural women’s education, Egypt 1992–2014

Source: Compiled from DHS surveys

10

30

50

70

90

Rural, Secondary > Urban, Secondary >

Rural, Literate Urban, Literate

20

40

60

80

100

120

140

160

Rural Urban

Source: Compiled from DHS surveys

90

skills, and growing unemployment and under-employment has been concentrated among high-school graduates, holders of secondary-level vocational qualifications, university graduates; as well among women (Fergany, 1995, quoted in Akkari, 2004).

The fate of rural girls who miss out on education is unenviable:

According to the Egypt Labour Market Panel Survey of 2006, 26 percent of girls aged 13–19 in rural Upper Egypt either received no schooling or dropped out after just one to two years. This means that more than 380,000 girls in the region are approaching adulthood without the skills and resources needed to succeed in a rapidly changing world. As girls reach adolescence, especially if they are not in school, community norms dictate that they should be closely supervised until a husband can be found. Arranged and early marriages are far more likely for these girls, followed rapidly by successive pregnancies, thus perpetuating the cycle of illiteracy and poverty into the next generation (Brady, 2007).

In response to this, a pilot programme, ‘Ishraq’ was launched to establish girl-friendly community spaces for girls to meet, learn, and play. Through a combination of literacy classes, life skills, and sports, Ishraq aims to provide rural girls with skills and more self-esteem (Brady 2007), By 2009, the programme covered 50 villages and 2,500 girls in Upper Egypt. A recent evaluation reports some positives, but queries the contribution to broader empowerment:

Although we find positive impacts on literacy, attitudes toward sports, and reproductive health knowledge, little impact was found on broader indicators of empowerment, and no impact on the attitudes of participants’ mothers or brothers (Sieverding and Elbadawy, 2016).

Women as citizens and agents, gender normsAlthough formally ‘government policy recognises the importance of women’s participation in all socio-economic and development activities and regards women as equal partners with men’ (National Strategy for Women, quoted in IFAD, 2006), in practice gender inequity is pronounced.

Property rights and inheritance. The rights of women of women to inherit and own property are guaranteed by the 1971 constitution, while the 1980 constitution gives women and men equal legal rights and obligations (Sonbol 2005). The 2014 constitution also upholds the equal status of women and men in ‘civil, political, economic, social and cultural rights’ (Government of Egypt, 2014). Despite formal protection, however, rights are not upheld owing to longstanding practice and weak implementation of formal law (Sonbol, 2005).

Prevailing practice follows Islamic custom whereby daughters inherit half as much as their brothers from their parents, wives inherit an eighth of a husband’s property, although a husband inherits all his wife’s property upon her death (Naguib and Lloyd, 1994; Cain et al., 1979; Amin and Lloyd, 2002).

While practice can be complicated, since there may be at least 20 different types of heirs, in general women get little. They may also be bought out by male relatives, or even have their rights usurped. These all contribute to the large disparities between male and female land tenure. In 2000 there were 8,417,962 male landowners compared to only 341,905 female landowners.

Moreover, changing customs to match formal law is not easy: forums for resolving disputes are largely controlled by men sympathetic to custom.

There are signs of progress, but they may take a long time to overcome resistance to changes to custom. The introduction of family courts and proposals to retain women judges in 2004 were lauded by activists (Sonbol, 2005), but do not seem to have achieved much. The current government has amended the law to punish those who deprive others of inheritance. A law to create an agency tasked with ensuring inheritances go to the right people (Al Monitor, 2016) has been drafted.

Violence and Female Genital Mutilation (Cutting) (FGM/C). FGM remains almost

91

universal in rural Egypt: 95% of rural women had been circumcised in 2014, a slight decline from more than 99% in 1995. The practice may be declining:

The prevalence of female genital mutilation and cutting (FGM/C) among girls aged 15–17 years is 77 per cent. The abandonment of FGM/C is a Government priority, and recent surveys indicate a lack of intent to circumcise girls in communities (UNICEF, 2016).

In 2014, almost 25% of rural women had experienced spousal violence: 43% of rural women agreed that a man would be justified in assaulting his wife, with failure to tell him she was going out being the most accepted justification (DHS survey).

Mobility norms. Women in rural Egypt are not allowed to move far outside the home without male supervision. For most, commuting or independent migration are out of the question. As much as any other factor, this contributes to the very low participation of rural women in paid work outside the household.

That said, Egypt shows that norms are inflexible. For many men, the rural ideal is that adult women remain within the house and compound and do not leave unless accompanied by a trusted man. From the 1970s onwards, however, when men left the villages for the urban employment, the women who remained took over their tasks in the fields – as described above. Although not ideal in the eyes of the men, the need to cultivate and harvest took precedence over norms.

4.4.4 Conclusion: women’s lives hemmed in by poverty and immobilityEgypt’s record of economic growth and transformation, in which agriculture has played a significant role, is not so dissimilar to that of Peru and Thailand. From the differences, however, come not only serious challenges to shared growth that reduces poverty, but also to rural women, who have made less progress than in the other two countries.

Three interrelated factors account for much the difference: the form of development and the creation of jobs; fertility and population growth; and, women’s mobility.

First, Egypt’s growth model has not created enough jobs, above all since the mid-1990s. It is not as though the government has not tried in the last 60 or so years: state-led industrialisation; redistribution of land to tenants and landless in the 1950s; and a succession of very large public construction projects, starting with the Aswan High Dam in the 1950s and continuing with desert irrigation schemes, were all considerable creators of jobs. But in recent decades, the employment elasticity of growth has fallen. Despite much economic liberalisation that might have been expected to attract foreign investors, Egypt has not seen the growth of assembly plants and light engineering in the ways that so many Asian emerging economies have. Moreover, much of the employment that has been created in Egypt has been regarded as male work, providing few opportunities for women. That tourism, in so many countries a major employer of women, should generate hardly any jobs for women in Egypt is remarkable.

Agricultural development has been largely successful in raising output ahead of growth, and creating jobs. But given the very high ratio of population to arable land, most additional jobs in rural Egypt have to be off-farm.

Second, Egypt is going through a prolonged, and possibly even stalled, demographic transition. Fertility rates, even in urban areas, have failed to fall toward natural replacement levels and remain stubbornly high. While this continues, Egypt faces the prospect of the labour force expanding by 2.9% a year: a rate that has easily outpaced the growth of jobs in most years since the mid-1990s. Not only is this condemning youth to a difficult search for employment that is often informal and ill paid; but also, it disproportionately affects women, and especially rural women who are less educated and less mobile than their urban counterparts.

Third, the lack of acceptance of women moving on their own severely curtails their job prospects, especially for rural women. Urban women may just find a job close to their home

92

where it is acceptable to make a short journey; that is much less likely in a village.

Rural girls thus face the prospect that their future lives will be restricted to a domestic domain where they will be expected to do much of the farm work, care for children, and dedicate their time to domestic chores. And to do so with few firm rights to property and under the threat of spousal violence.

All that said, poverty and deprivation are the still major determinants of life opportunities and options, more so than gender. To be born in a low-income rural household in Upper Egypt, perhaps landless, to parents of little education and few formal skills, means the chances of escaping poverty are low. This applies all the more so to girls, but the fate of the boys is not so enviable either. In such conditions, the search for gender equality goes hand in hand with poverty reduction.

4.5 Key messages from change in the long run

These three countries provide some common lessons about how countries can progress from low- to middle-income status, with structural transformation in which agriculture has a key role to play – and what this implies for the lives of rural women and girls. Four lessons stand out, as follows.

a) Broad-based development Economic growth broadly based by sector, urban and rural areas, and which tends to be labour-intensive can not only create much additional employment, but also jobs where average labour productivity rises as well.

Agricultural development is part and parcel of such growth and transformation. In the early stages of development of the three countries in the 1960s and 1970s, agriculture largely fulfilled the Johnston-Mellor functions of feeding the nation, exporting to obtain foreign exchange, raising rural incomes to provide domestic demand for the output of urban industry, and releasing labour and capital to the growing

34 If village social accounting matrices were available, we might run such counter-factuals, but the promise of those set out by Taylor and Adelman (1996) has never been realised.

urban economy. The exception here is Peru where agricultural development slowed well behind population growth from around 1967 to the early 1980s, stymied by policy biases against agriculture, under-investment in rural roads, and the unequal land tenure of the highlands.

It is not just agricultural development that counts in rural areas: the growth of the rural non-farm economy also matters. Since the RNFE is not a category in national accounts and is infrequently (unevenly and idiosyncratically) surveyed, we have less information on how it grows. All that can be seen for the most part is that surveys of rural households report large and increasing shares, often more than half of rural incomes, coming from sources other than farming. The few insights into the RNFE suggest that this sector grows primarily through two mechanisms: activities that are linked to agriculture either directly in the supply chains up- and down-stream of the farm or indirectly through spending by farm households; and public spending on rural infrastructure and services that create rural jobs. The exceptions to this come from Thailand, where putting-out has been used by manufacturers to take advantage of lower labour costs in rural areas – which can work for simple manufacturing that is labour-intensive; and where some assembly plants have moved from Bangkok and the Eastern Seaboard to the provinces – although they tend to establish themselves on the edges of provincial towns, rather than in villages.

Even though smallholder farming and the RNFE tend to be quite intensive in employment, rural–urban migration still plays an important role. Not only can this provide labour for an expanding urban economy, but it can also raise returns to household labour. A pertinent counter-factual is to imagine villages in all three countries, stripped of the assets that have been financed through remittances, shorn of the local businesses that depend on spending of such funds.34 It is likely that many villages would thus be much impoverished.

The generation of rural jobs, and migration options, seems especially important for rural

93

women. Within agriculture, women face two difficulties. One is that much agriculture has low labour productivity, even when using improved technology. The other is that women tend to find themselves either farming the less (labour) productive crops and livestock;35 or are deprived of access to inputs and technical innovations; or both of these. Compared to such work, a job in the RNFE or migrating to a city can pay so much more. Moreover, whereas while working within the farm household, a woman may see little cash income or see it appropriated by her husband, when working for wages or operating a small enterprise, a woman is more likely to control the income. And with that control can often come greater status and self-esteem.

A clear proposition emerges from these cases: getting out of agriculture matters more for rural women than for men.

b) Investing in human capitalTwo, rural people can only take up the better opportunities economic growth creates if they are healthy, literate and numerate. Hence public investments in rural education, health, and clean water and sanitation matter. Primary health services, water and sanitation, have done much to reduce disease: the decline in under-five mortality in all three countries, taking this a proxy for overall rural health, has been remarkable. In a generation, the pain of losing an infant has gone from being commonplace in villages to a tragic rarity.

Education provision has improved significantly in the rural areas of all three countries. Not necessarily smoothly so, as the delay before Thailand realised how far it was behind on secondary schooling shows. Questions remain about the quality of rural schooling; about equity of access – a prime concern in Egypt; and about the demand for secondary schooling for girls. In relation to demand, contemporary accounts from Peru suggest that a turning point has been reached. Faced by the realisation that the farm cannot provide living incomes for children, and the example of female employees in schools and

35 Labour productivity, not land productivity: women often tend home gardens, raise backyard poultry, and keep a couple of milk cows; activities that generate high yields per unit area – but some of which require hours of hard work.

health posts, rural parents are finally insisting that their daughters complete secondary schooling.

c) Family planningOne service matters more than expected before reviewing these cases: family planning. When formal contraception is readily accessible to rural women – which usually means services in the village, preferably on the door-step, and delivered by people they trust – then the uptake is rapid and strong. Adoption also tends to rise with girls going to secondary school as well.

Not only does contraception allow women to control their fertility, but also it leads to fewer births, and – with a delay of a decade or two – a pronounced slow-down in the growth of the labour force. Once rural populations cease to grow rapidly, so labour shortages arise, investment to raise labour productivity is encouraged, and wages for unskilled work in rural areas start to rise.

d) Gender normsGender norms count. While they are not set in stone, and subject to reinterpretation when it matters, norms can either facilitate or block these processes.

In these three countries, mobility stands out: rural women have many more options when it is acceptable for them to travel independently, to visit market centers, to commute for jobs, or to migrate to a city. Which is not to say that in societies where women’s rights are not firmly established in practice, there are no risks to women who travel. There are. But the point is to reduce the risk rather than prevent movement.

By and large, rural women’s lives have improved in all three countries, although notably less in Egypt. Rural girls face better prospects than their grandmothers did back in the 1960s and 1970s. Progress, however, is uneven.

Rural women have better options as workers and are likely to be better rewarded for their efforts. They are generally healthier – and less likely to have suffered from incapacitating malnutrition in infancy – and most have control over their fertility. Most have been to school, and increasingly

94

they will have been to secondary school. These are the clearest signs of progress seen.

Less clear are changes to norms by which the bulk of household chores fall to women, as well as almost all child care. Fewer children, piped water on premises and rural electricity may

36 Labour productivity, not land productivity: women often tend home gardens, raise backyard poultry, and keep a couple of milk cows; activities that generate high yields per unit area – but some of which require hours of hard work.

well have reduced some of the chores,36 but the gender disparity remains wide.

Progress on equity, whereby rural women are full citizens, the equal of men, is slow. Women’s status may be rising, but in rural areas women still usually continue to come second to men.

95

5 Changing lives of rural women and girls in contemporary Africa: Ethiopia and Tanzania

5.1 Introduction

Ethiopia and Tanzania share some characteristics. Both are low-income countries, still largely agrarian and rural, with natural resources that are mainly land and water, once abundant with respect to population but increasingly coming under pressure from relatively rapid population growth. Both had their economic growth and development interrupted in the 1970s by unwise policies. Subsequently, with changes of government and policy reforms, the two countries have since the early 1990s seen renewed growth with transformation.

Reviewing overall economic and agricultural development, this chapter looks at what growth has meant for the lives of rural women and girls in the two countries. The material conditions of existence for many rural women in both countries have improved considerably over the last 25 years. It is less clear that this has empowered rural women.

Given data limitations, it hard to make firm judgments about the degree of progress. That is all the more so when these cases are compared to those of Egypt, Peru and Thailand. In all three cases, but most disconcertingly for Thailand, a simple comparison of conditions for rural women in the mid-1980s with those in the early 1960s would not necessarily have given much insight into the conditions that would prevail by the mid-2010s. It is not that surprising that this

review of Ethiopia and Tanzania does not lead to firm conclusions about what the future holds for rural women and girls over the next few decades, say by 2035 or 2045. Hence this chapter concludes not by establishing a clear statement of progress over the last 25 years, but by identifying the factors most likely to make a difference for rural women and girls in the coming decades.

The rest of this chapter examines the two countries in turn and reviews, organised by the framework of overall economic growth, agricultural and rural development, changes in rural households and then the conditions of rural women as economic participants, as mothers, wives and carers, their human development, and as citizens and agents.

5.2 Ethiopia

5.2.1 Economic growth and structural transformation

Introduction: Ethiopia’s economic recoveryFrom the 1970s to the early 1990s Ethiopia saw much turbulence. In 1973 a severe famine led to the overthrow of the Emperor Haile Selassie by a military council, the Dreg, with socialist leanings.

The Derg redistributed land, nationalised enterprises and tried to direct and control the economy. Forced land resettlement and villagisation did much to disrupt agriculture, while conflicts with both Somalia and domestic

96

insurgents drained resources. The economy languished, discontent grew, until finally insurgents entered Addis Ababa in 1991 to overthrow the Derg.

The new government embarked on a more liberalised policy regime in 1992. A variety of market-based reforms were introduced to reverse the past policies, and a series of key policy documents that were primarily concerned with poverty and wellbeing in Ethiopia was issued. The Economic Reform Programmeme, implemented between 1992 and 1997 prioritised sound macroeconomic management, structural transformation that rewards both labour and capital, increased expenditure on education and health, enhanced food security, and deeper market integration with the overarching objective of reducing poverty (Guinand, forthcoming).

From 1994, the Agricultural Development-Led Development Industrialisation (ADLI) strategy was adopted. This aimed to stimulate farm output and rural incomes, through improved agricultural technology, roads, and credit, among other things. Not only would this boost farm output and rural incomes, but also through linkages this would boost domestic industry, starting with activities such as food processing (Ludi, 2004).

Economic recovery and growth Economic growth in the 1980s was set back by years in which a combination of bad weather and crop failure, economic disruption and conflict meant that GDP per person declined. Between 1981 and 1991, four out of ten years saw major setbacks: as result, GDP per person declined during the decade, reaching a low point of just US$164 (in 2010 terms) per person in 1992 (Figure 54).

Subsequently only two years have seen GDP per person decline.37 During the 1990s growth was modest and variable from year to year. Nevertheless, GDP per person had risen to reach US$197 by 2000. Since then, there has been accelerated growth: after the set-back of 2003, GDP per person has grown by 8–13% a year; so that by 2015 GDP per person was US$486, more than double that of 2000.

37 1998 saw war with Eritrea; 2003 was a drought year leading to lost harvests.

Growth during the 2000s has come from two main sources: agriculture and services (Figure 55). Given its large share of GDP, it is no surprise that agricultural growth matters so much for overall economic growth. Its contribution to growth, however, has declined, in line with agriculture’s falling share of GDP (Figure 56). Manufacturing and other industry have played only a small role in growth.

High economic growth has been driven by public investment. Disaggregating investment by ownership indicates that the share of public investment increased substantially from 45% of total investment in 2000/01 to 96% in 2010/11, while the share of private investment declined accordingly over the same period (Ferede and Kebede, 2015).

Structural changeWith economic growth has come a relative decline in agriculture: its share of GDP fell from 58% in 1981 to 41% in 2015 (Figure56). Services have taken over the largest share of GDP. Industry, of all kinds, has increased its share from 9% in 1981 to 16% in 2015; but manufacturing made up only a small fraction of the economy, 4%, in 2015.

Manufacturing is dominated by processing food, beverages, textiles, hides and skins, and leather. Growth rates have been comparatively slow, even if most foreign direct investment (FDI) has been in labour-intensive sectors and manufacturing, with tasks considered by employers to require female aptitudes (‘nimble fingers’) (Parker et al., 2016), thus potentially benefiting women substantially.

Manufacturing lacks skilled workers, access to land and electricity; while small and medium-scale enterprises have limited chances of obtaining formal credit. Manufacturing further faces an unfavourable tax system, poor trade logistics, customs procedures and trade regulations as well as cumbersome business regulations. Ethiopia is ranked by the World Bank 159 out of 190 countries in terms of ease of doing business (down from rank 146 of 189) (World Bank, 2017). Starting a business, accessing credit, protecting

97

minority investors and trading across borders are major areas of weakness. Furthermore, the absence of a level playing field between state-owned and party-affiliated firms and private firms is also hindering competition (Zerihun, Wakiaga, and Kibret, 2016).38

Employment and labour productivityTotal employment reached 40 million in 2013. The labour force is growing rapidly: 15 million

38 The Ethiopian government aims to address these constraints to the manufacturing sector through an ambitious Industrial Park Programme aimed at boosting the manufacturing industry via Foreign Direct Investments (FDI) (World Bank, 2015c).

were added from 1999 to 2013. Agriculture still provides most jobs: 73% in 2013, even if down from 89% in 1994 (Figure 57). Other major sources of work were services and construction. Manufacturing absorbed only a small additional portion of employment (World Bank, 2016).Labour productivity differences across sectors are wide. In 2010/11, employees in finance, utilities, transport and mining generated between US$3,750 and US$7,900; in manufacturing

Figure 54 Economic growth, Ethiopia, 1981–2015

Source: Compiled from World Development Indicators

-15

-10

-5

0

5

10

15

100

150

200

250

300

350

400

450

500

Annual GDP Growth (%

)GD

P pe

r ca

pita

(co

nsta

nt 2

010

US$)

GDP growth GDP per capita

Figure 55 Contributions to growth, Ethiopia, 2002–2012

Source: Compiled from Table 2, Ferede and Kebede (2015), based on MOFED data

0%

20%

40%

60%

80%

100%Agriculture Mfg Other industry Services

Figure 56 Sector shares of GDP, Ethiopia, 1981–2015

Source: Compiled from World Development Indicators

0%

20%

40%

60%

80%

100%

Agriculture Manufacturing Other Industry Services

98

US$575; and in agriculture, just US$375 per employee.39

5.2.2 Population changeEthiopia is the second most-populous country in Africa with an estimated population in 2015 of over 99 million. In the 1990s, population growth reached over 3.5% a year: since then growth has

39 Converting estimates in birr to US dollars at the December 2010 rate of ETB16 to US$1.

40 Urbanisation arises from a combination of natural growth (40% of increase), migration to urban centres (33% of increase), reclassification of villages to towns (24% of increase), and expansion of urban agglomerations largely to obtain land for industry and manufacturing (3% of increase) (Zerihun, Wakiaga, and Kibret, 2016). Despite its primacy, Addis Ababa has experienced a relatively low growth in economic activity compared with other cities in the country – especially smaller cities around the capital. Economic activity in other cities especially in the north of the country is declining, even over the 1994–2012 period, which has seen otherwise strong economic growth (World Bank, Cities Alliance, 2015).

slowed, but is still quite rapid: just below 2.5% a year by 2015 (Figure 58).

Although fertility rates have declined from more than 7 children per woman in 1990, they are still high, at 4.4 (see below for more on fertility).

An estimated 16.7 million persons lived in urban areas in 2014. At just 20%, urbanisation remains very low, compared to close to 40% for the rest of sub-Saharan Africa. The urban share is expected to increase considerably over the coming decades, since population growth in urban areas, at 5% a year,40 is greater than in rural areas, 2.5% a year (2015 figures). By 2037, more than 42 million may be living in cities. (World Bank, Cities Alliance, 2015).

Migration is quite low compared to other developing countries. Only 13.7% of the male population and 16.2% of the female population were migrants – defined as those living in a different district or city than the one they were born in – in 2012 (World Bank, 2015a), though the overall migration rate is increasing.

Three-quarters of migrants come from rural areas. Most, almost half of all migrants, have moved to other rural areas: they tend to come from poorer households. Those moving from rural to urban areas make up 25% of migrants: they tend to come from wealthier families.

Migrants are generally individuals, young and better educated than non-migrants. Most migrants (86%) move for work; 62% of them men. More women, however, migrate than men, but largely for marriage and other family obligations, rather than to work.

5.2.3 Agricultural and rural development

BackgroundExcept for the drylands where pastoralism predominates, most agriculture in Ethiopia has until recently been rainfed cultivation of staple

Figure 57 Shares of employment, Ethiopia, 1994–2013

Source: Compiled from World Development Indicators

0%

20%

40%

60%

80%

100%

Agriculture Industry Service

Figure 58 Population growth, Ethiopia, 1981–2015

Source: Compiled from World Development Indicators

2

2.5

3

3.5

4

30

40

50

60

70

80

90

100

Annual growth, %

Tota

l po

pula

tion,

mill

ions

Total population Annual population growth rate

99

crops by smallholders using few inputs other than labour assisted by draft animals, employing few tools other than plows, sickles, hoes and plows.

Production has been low, so that after meeting household needs for subsistence, only small surpluses have been produced, for sale and to build stores to cover years of poor harvests.

In arid and semi-arid areas, pastoralism is the main activity. Livestock are kept both for subsistence, and to accumulate herds that provide insurance against dry years and other shocks, and from which sales can be made to meet household needs.

Yields per unit area or head of livestock have been low, with meager labour returns that have left most smallholder households living in or close to poverty. Moreover, when the rains failed, famine threatened. Since the 1960s, famine struck in 1973 and again in 1984/85.41

Lack of progress on agricultural development can in large part be attributed to policies that either neglected agriculture, or actually hindered agricultural growth. In the 1960s, under the imperial regime, parts of the country had feudal land tenure, where peasants only had access to fields in return for labour and rents paid to the landlords (Githinji and Mersha, 2007). After the overthrow of the emperor in 1973, the incoming Derg regime reformed tenure by redistributing land from landlords to the tillers. While that might have been to the good, the government prevented the farmers from reallocating the land, while reserving the right to reassign it. Under the Derg, moreover, millions of rural households were moved into villages, no matter where their fields lay. Some were involuntarily resettled in distant lands. The uncertainties over tenure, and the inefficiencies of villagisation reduced incentives to invest and produce.

The socialist aims of the Derg saw attempts by the state to control the economy. In agriculture, the state Agricultural Marketing Corporation (AMC) demanded that farmers deliver quotas of grain, even to the point where farmers whose harvest had failed were expected to buy grain to meet the quota. Further controls applied to prices, while the exchange rate was overvalued.

41 Indeed, so vulnerable was Ethiopia to famine and food crises, that studies of rural Ethiopia before the 1990s were dominated by investigation of the causes of famine, and potential responses.

The result was that farmers were in effect taxed: by as much as 27% in 1992 (Figure 59) (Githinji and Mersha, 2007; Rashid and Negassa, 2011).

After the fall of the Derg in 1991, state controls were much relaxed and the exactions of the AMC ended. Moreover, the new strategies for economic recovery, growth and poverty reduction made agriculture the leading sector. Agriculture was thereafter to get support from agricultural extension services, investments in rural roads, education and health, and from a more stable macro-economy where private initiative, investment and innovation was enabled.

Agricultural growth and productivityFrom the early 1960s to early 1990s, agricultural output grew only slowly: on average by less than 1.5% a year, well behind population growth. Episodes of growth were interrupted by years of bad weather (Figure 60).

Subsequently, agriculture has grown strongly, with annual increases of more than 6% a year the norm. Production of all the main groups of crops rose between the early 1990s and mid-2010s, with similar increments to both area and yield per unit area (Figure 61). Land use, however, did not greatly change: more than 80% of the cropped area remained sown to cereals, roots and tubers, and pulses.

Figure 59 Net Rate of Assistance to Agriculture, Ethiopia, 1981–2005

Source: Anderson and Valenzuela (2007)

-30%

-25%

-20%

-15%

-10%

-5%

0%

100

For livestock, the largest increases have been seen in milk output.

Agriculture provides the bulk of the country’s exports: more than 80% of commodity exports between 2004/06 and 2013/14. Export performance reflects the changing fortunes of agriculture overall (Figure 62). From the mid-1970s to the early 1990s, the value of agricultural exports – in real terms – fell. After that came recovery interrupted by a slump in the late 1990s, with particularly strong growth in the

new century. Between 1993 and 2013, the value of farm exports rose more than eleven times in real terms.

Coffee is the most important export crop, accounting for 23% of the value of agricultural exports in 2013/14; its share much reduced from the 45% it represented in 2000/01. Growth of farm exports has thus seen some diversification. Export of live animals and flowers increased significantly – both were virtually absent in 2000/01, but now comprise of 6% each.

Figure 60 Agricultural growth, Ethiopia, 1962–2012

Note: Production until 1993 is for Ethiopia PDR; thereafter it is Ethiopia. The difference is that before 1993 these data include estimates for production from Eritrea. 3YMA = Three-year moving average to smooth annual variations.Source: Compiled from FAOSTAT data

-4.0

-2.0

0.0

2.0

4.0

6.0

8.0

10.0

2

3

4

5

6

7

8

9

10

11

12

Annual Growth, %

Gros

s Pr

odn,

US$

Int.,

200

4/06

con

stan

t

Annual Growth, 3YMA Gross Value Production, 3YMA

Figure 61 Increases in production, area and yield, main crops, Ethiopia, 1993/95 to 2012/14

Source: Compiled from FAOSTAT data

-0.5

0.5

1.5

2.5

3.5

4.5

Cereals (Rice Milled

Eqv)

Pulses Oilcrops Roots & Tubers

Fruit & Vegetables

Increased area Increased yield

Figure 62 Agricultural trade, Ethiopia, 1961–2013

Source: Compiled from FAOSTAT data, Current values in US$ deflated by US GDP deflator to constant 2002 values.

0

500

1,000

1,500

2,000

2,500

Valu

e, U

S$M

, con

stan

t 20

02

Exports Imports

101

Sources of agricultural growthAgricultural growth has come in part from added land and labour: between 1994 and 2013 the area under arable and permanent crops increased by 55% and the farm labour force expanded by 44%. The combination of these, however, does not fully account for an increase in output of 165% over this period. Productivity of both land and labour has risen significantly since the early 1990s (Figure 63).

For a country that for so long has had low yields per hectare of its staple crops, the yield increases are welcome. Given that most agricultural statistics come from estimates rather than measurements, however, one might wonder to what extent the increases are real. Fortunately, some corroboration can be found in reports from surveys. Bachewe et al. (2015) compiled estimates of average annual increases in cereals yields for periods since 1997 from different rural surveys. The great majority of estimates lie in a range from 4–6% a year (Table 3). The encouraging rise in reported cereal yields would thus seem to be real, rather than a statistical error.

In the new century, the government has shown increasing interest in the potential for accelerating agricultural growth through larger-scale commercial farming. Foreign investors have been welcomed. In regions with relatively abundant land, land has been leased to commercial investors – with some alleged dispossession of existing land users (Makki and Geisler, 2011). In 2010, it set up the Agricultural Transformation Agency (ATA) to facilitate investment and build links between public agencies, existing farmers, firms in the agricultural supply chains and incoming investors.

It is still too soon to see the results of this drive. At times, vast areas have been reported as allocated to investors with mention of more than 3M hectares. Recent estimates, however, lie in the range 1–1.5M hectares – of which perhaps as little as less than 40,000 hectares has been brought into cultivation (Ali et al., 2016).

One particular form of commercial farming, however, has certainly done well: floriculture. In the early 2000s both domestic and foreign investors set up flower farms in areas within easy reach of the international airport. Land was allocated to investors, credit was provided on generous terms through the Development

Table 3 Estimated annual increase in cereal yields, different sources, since 1997

Crop CSA Ad hoc surveys ERHS CSA CSA

Period 2002 2005-2014 2008-2013 1997-2012 2001-2012

Number of surveys Annual 8 surveys 2 surveys Annual Annual

Teff 5.8 4.7 1.7 4.2 4.6

Maize 6.2 6.2 0.4 3.4 4.2

Barley 4.8 6.8 10.1 - -

Wheat 5.4 6.3 3.6 4.0 3.6

Sorghum 5.4 -1.8 - 4.5 4.8

Source: Table 3.6 in Bachewe et al. (2015)

Figure 63 Land and Labor Productivity, Ethiopia, 1993–2014

Source: Compiled from FAOSTAT data. Land productivity is value of all agricultural output per hectare of arable and permanent crops. Since output value includes that of livestock grazed on pastures it overstates land productivity. Labor force surveys provide number of agricultural workers.

100

200

300

400

500

600

700

Vale

Pro

dn,

US$

04/0

6, p

er h

a Ar

able

&

Perm

/Wor

ker

Labour Land

102

Bank of Ethiopia, and the government subsidised freight rates on Ethiopian Airlines (Schaefer and Abebe, 2015). Flower farms use labour intensively: between 30,000 and 85,000 Ethiopians are estimated to be working directly or indirectly in floriculture. Some 60% to 70% of employees are women picking, sorting and packing the flowers. This work is, however, poorly paid, jobs are often insecure, working hours long and workers are exposed to hazardous chemicals (Kirigia et al., 2016).42

Drivers of agricultural growthA combination of factors has contributed to the strong increases in agricultural output seen.43

Immediate factors include increased use of better technology. Between 2004/05 and 2013/14, the fraction of farmers growing cereals using manufactured fertiliser rose from 46% to 76%: total fertiliser use in Ethiopia over the same period rose from under 300k tonnes to more than 700k tonnes.

More varieties of improved seed have been released since 2000, with some evidence that increasing numbers of farmers are using them – albeit from a very low base. By 2013/14 the highest reported incidence of use of improved seed was 28% for maize farmers: for all other cereals, the share of farmers using better seed was in single figures by that year.44

Official estimates show that less than 5% of arable and permanent crop land was irrigated by 2014. Nevertheless, use of irrigation is increasing: between 1993 and 2014, the irrigated area rose by more than two-and-a-half times. Moreover, such statistics may not fully count informal irrigation, especially that used in the off season.

Technical improvements have been facilitated by other factors that have allowed and encouraged farmers to intensify and expand

42 The flower farms have also been criticised for not having paid enough compensation for land allocated. The farms, however, have a small footprint: they are, to all intents and purposes, factories. Few spillovers arise from these farms to surrounding farmers. On the other hand, some farms provide health care and schools.

43 This section draws largely on Bachewe et al., 2015, unless otherwise referenced.

44 This, however, may understate genetic improvement. Field surveys of the provenance of varieties being grown show that improved strains may have spread further, especially – and as might be expected – with open pollinated varieties where farmers can multiply and spread improved varieties without necessarily buying fresh seed (Bachewe et al., 2015).

production. Bachewe et al. (2015) pick out four, as follows:

• Extension. More extension agents were posted to rural areas after 1992, but a major boost was made after 2002. The plan is that in every kebele – a ward, a sub-division of the district – there would be three advisors, each specialised in crops, livestock and natural resource management. By the late 2000s, 60,000 agents were in place, with one for every 476 farmers – a closer ratio than seen, for example, in China or Taiwan.

• Roads. In 1991, Ethiopia had just 5,610 km of rural road in a large country – about 5 km for every 1,000 km2. By 2000, there were 15,480 km of rural roads, and by 2008, 24,000 km (Rashid and Negassa, 2011). A more recent estimate for 2013 reports no less than 99,500 km of all-weather surfaced roads (Bachewe et al. 2015). That has been accompanied by an almost five-fold increase in the number of small trucks of 3–7 tonnes capacity between the 1990s and 2008 (Rashid and Negassa, 2011). Road expansion has dramatically reduced both the time and cost taken to reach urban centers. The percentage of people living within five hours’ travel to a city with a population of at least 50,000 rose from 33% in 1994 to almost 62% in 2007 (Chamberlin and Schmidt, 2011) (Figure 64). Between 1994 and 2008, the cost of moving cereals to market fell by more than 90% a tonne (Rashid and Negassa, 2011).

• Price incentives. Lower transport costs tend to raise output prices, and lower costs for purchased inputs, at the farm gate. Supply chains have become more efficient, with increased use of mobile phones, and with more storage in place (Rashid and Negassa, 2011). Prices for farm produce rose during the 2000s

103

in Ethiopia partly from rising international prices, and partly from expanding urban demand (Dorosh and Ahmed, 2009). An overvalued birr and controls on fertiliser prices have helped keep down costs of fertiliser to farmers.

• An expansion of rural education – see below – has led to a reduction in the share of farmers who are illiterate.

In addition, Ethiopia has been carrying out land certification to record farmers’ rights, first in Tigray beginning in 1998, then Amhara in 2003/04 and in Oromia and the Southern Nations regions from 2007 (Deininger et al., 2007; Bezabih et al., 2016). While the precise effects on output have yet to be established, it may well have contributed to the increased output seen.45

Assessing agricultural performance Since the early 1990s, the growth of Ethiopia’s agriculture has been quite exceptional, with rates to match some of the best-performing

45 Not only may certification confirm the rights of farmers and make them more secure in their tenure; but also in a country where the state has been prepared in recent history to interfere with farmers’ rights, it may help signal the government’s support for farmer initiatives.

agricultural sectors anywhere in the world. Given the low baseline, however, it has perhaps not been that hard to raise output.

Perhaps more remarkable is that much of the increase has come from smallholdings: the contribution of larger-scale farming has been small. That said, most of the increased output and intensification comes from the better endowed parts of the country, those with ready access to the cities, and within these areas, from farmers who have had the capital, labour and land to take up opportunities (Bevan et al., 2014).

Growth of the rural non-farm economySurveys in the 2010s show surprisingly little development of the rural non-farm economy (RNFE). In 2011/12, 19% of rural households reported being involved in any non-farm activity such as selling processed agricultural goods, providing non-agricultural services, trading goods or selling firewood and charcoal: this share was 18.4% in 2015/16 (CSA and World Bank, 2013; CSA, NBE, World Bank, 2017). Slightly higher participation was reported from a 2011

Figure 64 Travel time to reach a city of more than 50,000 persons, 1996/97 and 2010/11

Source: Kedir et al. (2015), reproduced in Bachewe et al. (2015)

104

rural survey: 25% of rural households with a non-farm business, 8% with a non-farm wage.46

Both the better-off and poorest households were more likely to engage in rural enterprises; but while the former tended to be trading livestock and consumer goods, the poorer households were engaged in petty activities, such as brewing and selling alcohol, selling food or making handicrafts (Bachewe et al., 2016). It seems that the poor are pushed into non-farm activities for lack of better options, rather than taking up attractive and productive work.

Clear differences in terms of the contribution of off-farm income to total income exist in terms of age and sex. Young household heads, especially young women, rely more on off-farm income than more mature household heads, largely because they do not own land at all or much less than more established households.

5.2.4 Poverty and inequalityAgricultural growth has resulted in broad-based reduction of poverty. Poverty incidence (headcount poverty index) decreased from 46% in 1995 to 30% in 2010, using the national poverty line47 (UNDP, 2015) and from 63% to 37%, using US$1.25 PPP per day as benchmark (World Bank, 2015a). Since 2005, 2.5 million people are no longer living in poverty. However, because of population growth, the absolute number of poor people has remained unchanged at 25 million over the past 15 years (UNDP, 2015).

Poverty declined in both rural and urban areas, though the rate of decline was higher in urban areas than in rural areas, largely owing to the boom in construction activities, increased private investment (e.g. in services), and expansion of small and medium-sized enterprises (SMEs) (Ferede and Kebede, 2015). Poverty reduction among rural, self-employed, agricultural households, however, accounted for the largest absolute number of people escaping poverty between 1996 and 2011.

46 Bezu et al. (2012) report higher participation, at 44% in 2004, a fraction that had – surprisingly – fallen since 1994. Differences in these statistics may well arise owing to the way that non-farm activities are defined are registered in different surveys. For example, how much time has to be spent on a non-farm activity – most weeks of the year, seasonal engagement, or any engagement at all – before the household is reported to participate in this?

47 The national poverty line is defined as 3,781 Birr per adult equivalent in 2011 prices, equivalent to approx. US$ 0.6/day

Structural change has not contributed much to poverty reduction during this time in contrast to some other economies in the SSA region and elsewhere (World Bank, 2015a).

Inequality among households in Ethiopia is low compared to other African countries: incomes may be low, but they are relatively equally distributed. In the 2000s and 2010s, however, inequality may have started to rise. At national level, consumption

Figure 65 Poverty headcount ratio at US$0.70/day, $1.25/day and $3.10/day 1995–2010

Source: World Development Indicators

0

20

40

60

80

100$0.70/day $1.25/day $3.10/day

Figure 66 Poverty headcount ratio using the national poverty line in rural and urban areas 1995–2010

Source: World Development Indicators

0

10

20

30

40

50

National Rural Urban

105

Box 8 A cautionary note on data

People who have worked in Ethiopia over the 20 years or so, and have spent time at the grassroots in villages or towns, sometimes find it hard to believe official figures, since their experience from talking to women and men and their observations sometimes contrast with what is being reported.

Questions about the reliability of official data have been raised by the IMF, stating that ‘deficiencies in source data persist which affect the overall estimation of national accounts’ (IMF, 2016) using the 2015/16 grain production data as an example. While the Central Statistical Agency (CSA) reported grain production to have grown by 2.9% in 2015/16, the FAO estimated crop losses at 14% and the US Department of Agriculture estimated overall losses in major grain crops at 20%.

The IMF also observes that ‘[…] national accounts statistics (especially from the expenditure side) remain work in progress, and economic growth, saving, and investment estimates are subject to considerable uncertainty—partly due to unreliable input data and survey methodologies’ (IMF, 2016).

Kebele staff, whether agricultural extension, health extension, nurses, teachers or kebele chairmen and the chairs of various committees are required to report regularly on progress in implementing the development programs and on achievement of targets. Compliance is sought through party discipline: civil servants and officials face regular gimgema1 where performance and achievement of targets are assessed. Civil servants face disciplinary action, for example demotion or posting to a remote location, if they do not report meeting or exceeding targets. Local civil servants thus have incentives to report improvements.

Because civil servants’ achievement of targets depends on what ordinary people do, there is widespread pressure to make people ‘volunteer’ in the various development programs. Beneficiary compliance is sought through incentives, persuasion in meetings, the threat of fines or imprisonment and/or the threat of exclusion from services, including the Productive Safety Net Programme (PSNP).

And finally, because of the tight control through official structures, including regular meetings and campaigns in which people are ‘awared’ about new policies and development programs, one-to-five household groups, and what is known as family police,2 some doubt remains, especially on sensitive topics such as female genital mutilation/cutting (FGM/C), how honestly people respond to questions.

Source: Own observations; (Bevan, Dom, and Pankhurst, 2013; Denney and Kassaye, 2013; IMF, 2016).

1 Gimgema: a forum for civil servants and party members to be lectured about new policies and development programs, a collective assessment of performance, target achievement of individuals and groups and individual behavior in relation to the group and the objectives of the organisation, and a forum where civil servants confess their shortcomings and achievements, which are then judged by their colleagues and superiors.

2 Family police: one family member, usually the household head, takes responsibility for resolving family conflicts. The family police representative must complete a form with the names and contact details of all members of the household and report any criminal or delinquent behavior up the community policing chain. Community Police Officers also make household visits to ask for reports from family police. This structure means that police have an ‘in’ to every household, which has serious ramifications for notions of domestic privacy.

106

inequality increased from a Gini coefficient of 0.29 in 2005 to 0.33 in 2011 (World Bank Data; Ferede and Kebede, 2015).

Long-term observations from 1995 to 2010 in 20 villages across Ethiopia representing different agro-ecological conditions, livelihoods, remoteness, poverty and market engagement (Bevan et al., 2010; Bevan et al., 2013) revealed that overall, communities were less poor in 2010 than they were in 1995. Major lifestyle changes happened over this period, especially for richer households who were often model farmers and therefore had better access to agricultural extension services, capital and modern inputs.

The observed general increase in wealth has been accompanied by increasing inequality among households living in the same village, with already richer households being able to accumulate wealth and engage in a broader range of livelihood activities, whereas poor households did not always receive the support they would have required. A new category of ‘very rich’ households emerged while at the same time poor households were increasingly differentiated into ‘poor’, ‘very poor’ and ‘destitute’. Inequality not only increased between richer and poorer households, but also between the older and the younger generations – youth unemployment, underemployment and landlessness have become serious concerns (Bevan et al., 2010; Bevan et al., 2013).

Between 1996 and the early 2000s, growth was broadly distributed, and the consumption of the poorest decile grew by more than the rest of the population. While the top 60% still experienced an average annual growth in consumption of 1.1% between 2005 and 2011, the bottom 10% saw an annual decrease in consumption of 1.9% (World Bank, 2015a) (Figure 67). Food price inflation, which peaked in 2008, 2009 and 2011 and reached at least 39% a year, led to reduced consumption for poor households who depend on markets to complement their own production, while it benefitted households that were net producers. The situation was exacerbated by a decline

48 The Multidimensional Poverty Index (MPI) is based on 10 indicators across 3 dimensions: Education: Years of schooling, school attendance; Health: Child mortality, nutrition; Living standard: Electricity, sanitation, drinking water, floor, cooking fuel, asset ownership (Alkire and Santos, 2010).

in rural real wages in 2008 and in 2010/11 (Bachewe et al., 2016).

While income-poverty has declined, the share of rural people living in multidimensional poverty – a combination of education, health and living conditions48 – fell only slightly from 99% in 2000 to 94.9% in 2011, despite significant investments in social services such as education, health and water supply and overall GDP growth (Figure 8). The absolute number of people living in multidimensional poverty increased in rural areas from 56 million to 69 million over the

Figure 67 Consumption growth for the bottom 10%, bottom 40% and top 60% 1996–2011

Source: World Bank, 2015a

1

2

34

0

-2

-1

-3

Bottom 10% Bottom 40% Bottom 60%

Figure 68 Rural and urban multidimensional poverty 2000–2011

Source: Alkire and Santos, 2010; Oxford Poverty and Human Development Initiative, 2013

0

20

40

60

80

100

0

0.2

0.4

0.6

0.8

2000

2005

2011

Percent Population

Mul

tidim

ensi

onm

al P

over

ty In

dex

(MPI

)Headcount urban Headcount rural

Headcount national MPI urban

MPI rural MPI national

107

same period (Alkire and Santos, 2010; Oxford Poverty and Human Development Initiative, 2013). The provision of services in rural areas and in deprived parts of urban settlements as well as general increases in income could not keep up with the increase in population.

The slow pace of structural transformation has prevented the creation of sufficient income-earning opportunities, especially for poor people in rural areas. Limited livelihood opportunities outside agriculture, including through migration, constrains strategies available to households to escape poverty and cope with shocks, especially problematic in an agricultural sector that is highly dependent on climatic conditions and which is increasingly affected by climatic change.49

5.2.5 Gendered aspects of transformation

Women as economic participants

Women’s economic participationAlthough women’s economic participation in Ethiopia is higher than in other countries, it is still significantly lower than men’s. For example, in 2014 the share of those aged 15 and above who were employed was 89.3% for men and 78.2% for women (Beyene, 2015). Women are also more likely to be employed in the informal economy and are thus more vulnerable to unemployment or redundancy and poverty. Women are particularly affected by unemployment. Nationally, unemployment stood at 16% in 2013. Unemployment among men at 10.5% was much lower than for women at 23%. Women aged 20–24 years had the highest unemployment rate at 30.2%, while men in the same age cohort experienced an unemployment rate of 23% (Beyene, 2015). Reasons for women not to participate in productive work were education, pregnancy, caring for children, and domestic duties (ibid.).

A gender wage gap exists across all sectors in Ethiopia, whereby men are paid more than

49 Since the mid-1970s, most of the country has seen increased warming and drying and parts of the country has seen considerable declines in spring and summer rains (Funk, et al., 2012).

50 Although the constitution grants equal access to land for men and women, women were allocated less than men (Girma and Giovarell, 2013), resulting in overall smaller landholdings of female-headed households whose holdings averaged 0.79ha, compared to the average of 1.07ha for male-headed households (CSA and World Bank, 2015).

women for the same job. In the formal economy, the lowest paid jobs are in the agricultural and food services sectors. Here approximately 547,000 men are officially employed and 142,000 women, with men earning roughly 733 Ethiopian Birr a month and women 561 Birr (Beyene, 2015). Although still low, real wages rose considerably from 2005 to 2010 for both men and women (Figure 69), though far more for men, whose wages were almost five times higher in 2010 than they were in 2005, while women’s wages increased by a factor of 2.5.

Women in agricultureDespite important changes to recent law and the recognition of equal rights for women and men in Ethiopia’s constitution, these policies have not necessarily translated to the local levels, especially in agriculture (Warner et al., 2015).

A significant productivity gap persists in self-employed agriculture between men and women (largely female-headed households) (World Bank, 2015a; Palacios-Lopez et al., 2017). Almost half of the productivity difference between men and women can be explained by differences in control of factors such as land area50 and quality, tenure

Figure 69 Average real wage from paid employment per sector and for women and men

Source: ILO

0

500

1000

1500

2000

Agric

ultu

re

Indu

stry

Serv

ices

Mal

e

Fem

aleCo

nsta

nt 2

000

Birr

/mon

th

2005 2010

108

security, access and use of irrigation, access to oxen, access and use of fertilisers, seeds, other inputs, access to extension services and labour input to farming (World Bank, 2015a).

When female and male farm managers were compared, after matching by age, crops produced, agro-ecological region, number of household members and household dependency ratio,51 almost no difference in returns to productive factors could be seen. Matched male managers were 17% more productive than female managers, but almost all of this difference (97%) could be explained by differing access to productive resources. (World Bank, 2015a).

On farms, women tend to be given low-status and low-paid (or unpaid) jobs, such as weeding. Gender norms see women as unable to carry out tasks such as lifting or carrying heavy loads – although this does not apply to carrying water and firewood. Women still do at least a third of the labour required for agricultural production in Ethiopia’s households (Palacios-Lopez et al., 2017).

Moreover, women plowing with oxen is typically considered taboo (Gella and Tadele, 2014; Alesina et al., 2013; Lavers, 2017). Since they are not allowed to use a plow themselves, the 23% of rural households headed by females in 2011 (CSA and ICF, 2012) have to hire male relatives or labourers to do the work, which is then often done late and so reduces yields. Alternatively, they sharecrop out their land, in return for one third to a half of the crop.

Women’s disadvantages as farmers are compounded by the neglect of their circumstances by research and extension. Agricultural extension services have focused on men and on improving productivity of cereals and livestock, which are generally seen as men’s domain, while services for women have focused on their traditional domains such as poultry and backyard vegetable gardens (Gella and

51 In the sample used, over 95% of female farm managers resided in female-headed households, and 99% of male farm managers resided in male-headed households (World Bank, 2015a). Based on the Agricultural Sample Survey, covering close to 47,000 holders, more than 19% were women, of which over 96% resided in female-headed households (Kasa et al., 2015). The share of rural households headed by women was 20% in 2005 and 23% in 2012, according to DHS data (CSA, ORC Macro, 2006; CSA and ICF, 2012). Some of the most productive male plot managers could not be matched to any women based on their characteristics; while some female plot managers comprising the least productive could not be compared to any male manager based on their characteristics.

Tadele 2014). Women receive far fewer visits by extension agents than men, in a 1:50 ratio, and women are rarely model households or participate in on-farm experiments (Mulema and Damtew, 2016; USAID, 2016). Technologies and interventions promoted, including investments in natural resource conservation, preparation and use of compost or row planting, generally increase the workload and therefore do not address women’s needs for time and labour-saving technologies.

Women in male-headed households face additional challenges as agricultural knowledge is poorly transferred from husbands to wives and other female members. Furthermore, women are often not allowed to travel if extension events are held outside the village (Lemma and Tesema, 2016; Mekonnen and Asrese, 2014).

Women face more barriers than men in obtaining access to credit, though this varies by region. Female-headed households face considerably more difficulties: in one study 58% of male-headed households received credit from a cooperative compared to 38% of female-headed households. The latter, on the other hand, were more likely to access credit from the Ministry of Agriculture. Overall, 21% of male-headed and 17% of female-headed households accessed inputs on credit (Minot and Mekonnen, 2012).

Microfinance institutions (MFIs) are another source of credit, though they are few and have limited reach in rural areas because of infrastructural problems such as breakdown in mobile phone connectivity and energy, and the tendency of commercial banks to provide loans to larger and established enterprises but not to small and medium-size enterprises (SMEs), new and young enterprises (Alemu, 2014). In addition to the more formal MFIs, cooperatives and credit unions provide access to finance for rural people, but cannot meet the credit needs of

109

their members from the resources mobilised from members’ savings.

Participation in rural savings and credit cooperative organisations (RUSACCOs) is often seen as a means of empowering women psychologically, socially, economically and politically (Brody et al., 2015). RUSACCOs and self-help groups also aim to build women’s leadership skills, provide financial literacy and entrepreneurship training and improve women’s political participation. Overall, such interventions are aimed at creating a more supportive environment for women to have control over income and to increase women’s self-confidence to access resources (Johnson et al., 2017). Participation in RUSACCOs has improved women’s access to resources, involvement in major household decision-making, the perception of their legal position and mobility (Rani and Yadeta, 2016).

Women in the rural non-farm economyBetter education, smaller families and some improvements in domestic labour-saving technologies have enabled some women to take up jobs outside the home. Nationally, women’s employment has risen, especially in the service sector. Between 2005 and 2013, the share of women employed in services increased from 16% to 27%, while the share of women employed in agriculture declined from 75% to 65% (World Development Indicators).

Women as migrantsBecause income opportunities in rural areas are still limited and service jobs where women are finding employment are concentrated in towns and cities, and because women also migrate for non-work reasons such as marriage and other family obligations, migration rates among women are higher than among men, though still low compared to sub-Saharan Africa. In 2012, only 16.2% of the female and 13.7% of the male population migrated (World Bank, 2015a). Generally, employment opportunities were the main pull factors for migrants, especially for men, whereas more women migrated for other reasons such as marriage and other family obligations or to escape early marriage or sexual abuse. Migrants are generally young and better

educated than non-migrants, though women migrants in Addis Ababa are ten times less literate than non-migrant women (World Bank, 2010) and face new challenges and often end up in very vulnerable positions and exposed to exploitation and abuse.

Rural–urban migrants come from wealthier families, whereas rural–rural migrants come from poorer households than non-migrants. When households with small land holdings manage to increase agricultural income, it becomes more likely that one member will migrate. Increasing income from agriculture may contribute to alleviating credit constraints that would otherwise limit migration (World Bank, 2015a). Women mentioned land scarcity, lack of alternative livelihoods, limited economic opportunities in rural areas and inadequate infrastructure as well as seasonality of labour as encouraging migration to more developed cities (Atnafu et al., 2014). Migrating to urban areas also exposes women to more gender-egalitarian practices (Marcus and Harper, 2015) and can play an important role in changing gendered cultural norms.

Despite increasing migration, remittances are still relatively uncommon in rural Ethiopia. Most migration is internal and is often an unplanned event for the poorest individuals and households rather than a strategy to escape poverty. In addition, wages paid, especially in those sectors open to women – domestic services, low-skilled agricultural work and manufacturing – are very low and migrants might even depend on support from their families to cover living costs in towns instead of sending remittances to them.

One emerging issue in Ethiopia is the increasing number of women domestic workers and commercial sex workers who migrate to other countries in search of employment. From 2008/09 to 2012/13, the total number of legal migrants reached 460,000, of whom almost 95% were women travelling to Saudi Arabia, Kuwait, Dubai, Djibouti, Yemen or Sudan (UN Women, 2014).

Possible reasons for why migration in Ethiopia is comparatively low relate to policies that favour equitable land distribution and rural development, e.g. measures to improve productivity, investments in agricultural extension, and the provision of the PSNP in

110

food-insecure areas, which have contributed to promoting growth in rural areas. The land registration and certification programme, launched in 1998, improved land-use rights and land security, which has helped to maintain an equitable land distribution in rural areas dominated by smallholder farmers, but because land sales are prohibited, which has prevented people from selling land and migrating to urban or other rural areas in search for more rewarding jobs, thus hindering a more rapid structural transformation and the associated gains in reducing poverty.

Women as mothers, wives and carersEthiopia’s rural society is characterised by a deeply engrained gendered division of labour which is changing only slowly. Women’s roles tend to be domestic, with the burden of household chores, care work and unpaid community work falling to them (Beyene, 2015), in addition to doing at least a third of the agricultural labour (Palacios-Lopez et al. 2017). Men do far less household work, which allows them to focus on agricultural production (Warner et al., 2015), taking up public roles and leadership positions in the community (Holden and Tefera, 2008), or just enjoy more leisure time.

Although women provide the vast majority of the unpaid care economy in all countries, in Ethiopia it is particularly marked: 88% of male respondents reported spending no time at all on care as a primary activity (Karimli et al., 2016). It is not only mothers who carry this burden: unpaid care is also provided by adolescent and even younger girls. Evidence from parts of Ethiopia suggests that 52% of rural girls between the ages of five and eight are engaged in care work, compared to 38% of rural boys. One-quarter of these young girls spend three or more hours daily on unpaid care (Samman et al., 2016).

It is thus not surprising that it is largely women and girls who are responsible for household chores such as preparing food,

52 Some debate remains about coverage figures. The National WASH Inventory was the country’s first comprehensive audit of water supply and sanitation systems, and classified ‘at least’ 25% of systems as non-functional. Official 2011 WASH coverage figures were revised sharply downwards; for rural water supply, from 66% to 49% (Butterworth et al., 2013). Subsequently, both JMP and national sector estimates have risen, with the JMP reporting 57% overall coverage (the Millennium Development Goal (MDG) target level) and 49% rural coverage for 2015 (Unicef and WHO, 2015).

fetching water and collecting firewood. Water and firewood collection, in particular, involve a heavy work burden and can take several hours a day. Food preparation using simple wood stoves carries significant health risks from indoor pollution, particularly affecting children and women.

In 2015/16, 54% of female household members spent time daily collecting water and fuelwood; while only 22% of male household members were engaged in these activities. This gender disparity is similar across all regions, rural areas, small and large towns. On the other hand, men are more likely than women to be engaged in agricultural activities: in 2015, 59% of men were performing agricultural work in the previous week compared to 41% of women (CSA, NBE, and World Bank, 2017).

While in 1990, over 90% of the rural population relied on surface water or unimproved sources for drinking water, this share declined to 51% in 2015 (Unicef and WHO, 2015).52 (Figure 70) Better access to improved sources for drinking water (such as dug wells fitted with hand pumps, protected springs or boreholes) resulted in health and nutritional benefits (reduced incidences of diarrhea and other water-borne diseases), plus

Figure 70 Drinking water sources, rural Ethiopia, 1990 and 2015

Source: UNICEF, WHO, 2015

134843

3554

16

0

20

40

60

80

100

PERC

ENT

Piped on premise Other improved

Unimproved Surface water

111

reduced exposure to physical hardship and sexual and physical violence associated with travelling long distances.

Persistent concerns remain, however, about the level of services people actually receive in terms of quantity, quality, reliability and accessibility (Calow, pers. comm.). There is some evidence that improved access to safe water has freed women and girls’ time, but there was no change in the percentage of households reporting spending more than 30 minutes for a round-trip to collect water (52.1% in 2005 and 52.6% in 2016) (DHS). There has also been no change in who is responsible for collecting water: in rural areas it is still mostly women and girls (81% in 2016) (DHS).

A similar picture emerges regarding firewood collection, where women are responsible for 70% of the work. The slightly higher share of male involvement in comparison to fetching water and the reported reduction in time spent to collect firewood from 0.6 h/day to 0.2 h/day ESRS/LSMS, 2011/12 and 2013/14) is possibly a result of planting more trees around homesteads, some of which is also used for fuelwood.

Women’s human developmentToday, women and girls have far greater chances than ever before to attend school and lead a healthier life. As set out in the various five-year development and growth and transformation plans, the government and its development partners emphasised the provision of basic services, especially access to transport infrastructure (rural roads), preventive and curative health services, improved drinking water and sanitation, and education. Many of these interventions have particularly benefitted women and girls and resulted in positive outcomes in terms of their health, educational attainment, time spent on household chores and engagement in economic activities.

EducationAchievements in providing primary education have been particularly impressive. While Ethiopia had one of the world’s lowest enrolment rates in the 1980s and early 1990s, net primary enrolment in 2012/13 was estimated at 86% (Lenhardt et al., 2015).

For primary education, the situation for girls has improved significantly. Girls have benefitted even more than boys from the expansion of access to primary education: the ratio of girls to boys enrolled has increased from 0.66 in 1991 to 0.94 in 2012/13 and girls now outnumber boys in lower secondary school (grades 9 and 10) (Ethiopian Federal Ministry of Education (EFME, 2013, cited in Lenhardt et al., 2015).

Secondary school attendance, however is low, very low in rural areas, and has actually been falling since 2000 (Figure 72). In 2011, only 6% of rural girls were attending secondary school. Attendance at rural secondary schools lags well behind that in urban areas.

It is difficult to encourage more rural girls to attend secondary school because so few rural households live close to one: more than 75% of rural households live more than 10 km away from the nearest secondary school. Lack of transport makes these schools inaccessible, and few parents are able or willing to rent accommodation for their children close to secondary schools. Moreover, once girls reach puberty and start menstruation, the lack of access to safe water and sanitation facilities at schools and provisions for menstrual hygiene can act as a strong disincentive for girls to enroll in education or lead them to drop out (Jasper et al., 2012). Norms that favour early marriage further discourage girls’ secondary schooling.

Even if girls today are more likely than their mothers to have been to primary school, the overall picture is still bleak. Between 2000 and 2011, average years of women’s schooling increased from only 1.1 to 2.9 years; with rural women averaging just 1.8 years of schooling while urban women had an average of 6.4 years in 2011 (Lenhardt et al., 2015). Overall, less than 35% of women aged 15 to 24 years completed primary education and less than 6% of women aged 20–49 completed secondary or higher education (Head et al., 2014).

Improved access to schools has not been matched by similar improvements in attainment. Absenteeism is another problem with up to half of all pupils not attending school (Tafere and Pankhurst, 2015). Many see the rapid expansion of enrolment to have come at the cost of quality,

112

so that education outcomes have declined in recent years (Lenhardt et al., 2015).

HealthSince 1993, policies to address health and population have been emphasised. They have focused on increasing access to family planning, improving maternal health, reducing infant and child mortality, improving access to safe water and sanitation, and improving education (Admassie et al., 2015).

One of the cornerstones of the push toward preventive care has been the launch of an extensive programme to deploy health extension workers (HEWs) to health posts. By 2013, there were more than 16,000 health posts with more

than 34,000 HEWs, many of them women (Figure 73). They play a crucial role in hygiene and sanitation education, the promotion of anti-malaria bed nets, the provision of basic mother and child health services and increasing access to modern contraception (EPHI, 2013) (Figure 74).

Investments in primary health have paid off: life expectancy has risen substantially and was estimated in 2014 at 66 years for women, up from 40 in 1960, and 62 years for men, up from 37 in 1960 (World Bank Data). Many more children were immunised in 2014 than in 2000. Stunting, while still high at 40% and concentrated in rural areas, has decreased, especially among better educated and wealthier families (CSA and ICF, 2016).

Between 2000 and 2016, the under-five mortality rate decreased from 166 deaths per 1,000 births to 67 deaths per 1,000 births, though big regional variations remain (Admassie et al. 2015). Maternal mortality, while still among the world’s highest, (Figure 75) more than halved between 1993 and 2000 and 2009–2016 from 871 to 412 per 100,000 live births (CSA and ICF 2016).

While overall very positive results have been achieved, it was also reported that infrastructure, furniture, and equipment were lacking, that health posts lacked access to electricity and water, and that vaccinations, contraceptives, basic medicines and drugs were provided only intermittently (Bevan et al. 2010) and 88% of women reported facing barriers to accessing health care (Head et al., 2014).

Fertility and family planningBefore the mid-1990s, on average, a woman had seven or more children. By the mid-2010s, the average fertility rate had fallen to below 4.5: lower than average for sub-Saharan Africa. In rural areas, however, on average a woman still has more than five children. Much of the decline in the fertility rate is attributed to the fast expansion of primary health services and the deployment of HEWs, including an extension programme on modern family planning and the provision of free contraceptives, reductions in infant and child mortality, a drop in child marriage and early childbearing, and improvements in women’s education (Admassie

Figure 71 Net attendance ratio, primary schools, Ethiopia, 2000–2011

Source: DHS

0

20

40

60

80

100

Male Female Male Female

Urban Rural

2000 2005 2011

Figure 72 Net attendance, secondary school, Ethiopia, 2000–2011

Source: DHS

0

20

40

60

80

100

Male Female Male Female

Urban Rural

2000 2005 2011

113

Figure 73 Human resources and health infrastructure, Ethiopia, 2001–2013

Source: EPHI, 2013

Figure 74 Coverage of key reproductive, maternal, new-born and child health outcomes, Ethiopia, 2000–2014

Source: EPHI, 2013

et al., 2015). In addition, there have been strong government and NGO campaigns on the economic and social benefits of smaller families.

The availability and use of contraceptives has increased greatly over the past decades. Today, 35% of married women are using a form of modern contraceptive; this was only 6% in 2000. There are still major differences across regions, ranging from less than 2% in Somali Region to over 50% in Addis Ababa. While only 30% of women without education use contraceptives, this increases to over 50% among women

with secondary or tertiary education. Similarly, poorer women are less likely to use contraceptives: 22% among the poorest wealth quintile compared to 46% among the wealthiest quintile (CSA and ICF, 2016).

Increased use of contraceptives was supported by social networks and social interactions, large family size and more boys surviving, which might indicate that contraceptive uptake is influenced by competition over resources, especially land, and the need for higher investment per child in education as the economy undergoes transition

Figure 75 Maternal mortality ratio (MMR) with confidence intervals, Ethiopia, for the 7 years preceding 2000, 2005, 2011, and 2016

Source: CSA and ICF, 2016

200

400

600

800

1000

1200

Seven years preceding the 2000

EDHS (1993-2000)

Seven years preceding the 2005

EDHS (1998-2005)

Seven years preceding the 2011

EDHS (2004-2011)

Seven years preceding the 2005

EDHS (2009-2016)

1039

703

548

799

541

810

551

273

871

673 676

412

1888 2668488

5261

11976

45509

862

5774

0

34382 115

150

1362

3315

0

16251

Doctors Health Officers

Nurses Midwives Health Extension Workers

Hopsitals

Health Centres

Health Posts

Human Resources Health Care Facilities

01020304050607080

2000 2014

114

(Alvergne et al., 2011). Despite the much greater use of contraception, more could be done: at present, only 62% of the potential for family planning is being met. More women would want to space or have fewer children if had they access to modern contraception.

Health committees have been established as part of the overall organisation of communities toward playing an active role in the various government-initiated development programmes. There have also been also attempts to organise women into women’s development teams and one-to-five groups to promote awareness of mother and child health, family planning and harmful traditional practices. Model families have been identified to serve as living examples of the HEWs’ extension messages. However, these structures have apparently not been very successful (Bevan et al., 2013).

Social ProtectionIn 2005, the government and donors launched the New Coalition for Food Security, including its Food Security Programmeme, consisting of four pillars:

• the Productive Safety Net Programmeme (PSNP) for chronically food-insecure households, providing food or cash transfers on a regular and predictable basis in exchange for work

• provision of agricultural and financial services to the poor through the Household Asset Building Programmeme (HABP)

• resettlement of families from land suffering erosion and loss of soil fertility

• provision of critical community infrastructure (Sandford and Hobson, 2011).

The PSNP was particularly important for female-headed households, which were twice as likely to take part compared to male-headed ones (Mariotti and Diwakar, 2016). This higher participation was facilitated by design, which recognised the particular vulnerabilities of women and female-headed households who are often more labour-poor than other households, have lower physical labour capacities, are time constrained due to domestic work and care responsibilities and face specific factors such as pregnancy (Holmes and Jones, 2011).

Insufficient attention was paid, however, to promoting women’s meaningful participation in the programmes beyond a focus on numbers and limited emphasis on addressing unequal gender relations in food security and agricultural productivity at the household and community level (focus on ‘practical gender needs’ rather than ‘strategic gender needs’) (Holmes and Jones, 2011).

Overall, the PSNP has helped meet several of women’s practical needs. It has helped to increase

Figure 76 Total fertility rate, Ethiopia, 1990–2014

Source: World Development Indicators, DHS

7.25 7.006.53

5.69

4.904.40

012345678

Birth

s pe

r w

oman

Ethiopia Sub-Saharan Africa

0

2

4

6

Num

ber

of c

hild

ren

per

wom

an

rural urban

115

household food consumption and contributed to meeting household costs. It has resulted in fewer distress sales of household assets and improved access to credit. Despite low payment, in some areas teenage girls and young women mentioned that participating in the PSNP had reduced their need to become domestic workers in nearby towns, which is paid even less and where abuse by employers is widespread. Some women also reported that working alongside men has helped increase their social status and respect within the household and community, even though it has not translated into changes in household decision-making. Some of the investments, such as the creation of water-harvesting facilities, have also benefited women directly and eased their domestic chores. Finally, having regular cash or food transfers has helped households to gain social capital through participation in social networks (Holmes and Jones, 2011).

The way the programmes are implemented, however, reinforces gendered norms and notions of work regarding food security and agriculture rather than dismantling them. The impact on unequal gender relations within households and communities has therefore been very limited. Despite formal provisions for equal pay, men were often paid more as their labour was more highly valued, women’s time poverty was not addressed to any significant extent and women found it difficult to balance participation in public works programmes with household responsibilities, investments that contributed directly to reduce women’s heavy work burden were rarely prioritised, and payments from the PSNP were made to the male head of the

53 Since 2003, the government embarked on an extensive decentralisation programmeme. Kebeles became the main arena where development interventions were implemented and a range of frontline government staff were posted. Following a ‘developmental state’ approach to state building, government and party structures became increasingly intertwined.

Communities were organised on four levels: kebele, sub-kebeles (2 or more per Kebele, depending on size), Development Teams (18-40 members) and 1-5s comprising farmers who shared borders. 1-5 leaders are often model farmers. The approach is meant to ensure the smooth flow of information from Wereda experts to Kebele staff, who would train Development Team leaders who in turn would train 1-5 leaders who are then passing on the learning to the members in their teams. This same structure is also used to report back on implementation progress and achievement of targets (Bevan et al., 2013).

Especially since GTP I, the implementation of development initiatives was shifted towards Development Teams and 1-5s. These structures were designed to involve all households in development, to establish community-based organisations (CBOs) and to involve elders, clan leaders and church and Muslim leaders. A considerable amount of time was spent ‘awaring’ people of the various development programmes and initiatives in government and party meetings. Interventions involved the use of packages, model households, quotas and targets.

household, even if women and children did the bulk of the work. It thus did not contribute in any meaningful way to women’s economic empowerment or household decision-making dynamics (Holmes and Jones, 2011; World Bank, 2013; Berhane et al., 2013).

Women as citizens and agents, gender norms Since the mid-1990s, women have become more involved in economic activities. This was supported by the reform of the family law in 2000 and better implementation of the right to land (see below), especially in recent years. Physical security has improved for women and girls and harmful practices such as FGM/C, abduction, early marriage and widow inheritance have diminished. However, women’s participation in in various development-focused committees and in political affairs has remained very limited.53 Patriarchy is still deeply rooted and an ideal household is one that is headed by a man, although men’s authority over women and youth has declined, partly as a result of government-led information campaigns promoting knowledge of women’s rights, better education and the increasing availability of information through various media also in remote villages (Bevan et al., 2013).

A range of awareness-raising campaigns and development programmes and packages was focused directly on women or aimed at changing perceptions of women’s economic, social, cultural and political roles. Overall, women have become more economically active, more aware of their rights and more assertive, but this has not necessarily led to greater

116

political participation or changed gendered norms and roles guiding women’s and men’s behavior (Watson and Harper, 2016).

Women and youth organisations have been formed at Kebele level, but they were often not effective as they had insufficient support at Kebele and Woreda level and lacked broad-based participation – women had other work to do at home. Successful development initiatives with a specific focus on women were in the health sector, whereas few livelihood-focused interventions for women, including women’s producer cooperatives, were sustained for any length of time (Bevan et al., 2013).

Legal reformsEthiopia’s constitution of 1995 (FDRE 1995) and subsequent Regional constitutions grant equal rights to men and women.54 Because of the historical legacy of inequality and discrimination, the constitution also grants women ‘affirmative measures’ with the aim of enabling women to participate equally with men in political, social and economic life as well as in public and private institutions. Article 35 deals with rights of women and states that

Women shall, in the enjoyment of rights and protections provided for by this Constitution, have equal right with men.

The State shall further

enforce the right of women to eliminate the influences of harmful customs. Laws, customs and practices that oppress or cause bodily or mental harm to women are prohibited.

The Constitution also ensures equal rights to women to access, administer, own and transfer property, including land (FDRE, 1995).

Two areas of legal reform have had a positive impact on women’s lives and contributed to their economic, social and cultural empowerment,

54 ‘All persons are equal before the law […]. […] the law shall guarantee to all persons equal and effective protection without discrimination on grounds of […], sex, […]’ (Art. 25) and ‘Men and women, without discrimination […] have the right to marry and found a family. They have equal rights while entering into, during marriage and at the time of divorce. […].’ (Art. 34) (FDRE, 1995).

participation in the labour force and uptake of on- and off-farm employment: changes to the Family Law and a land-titling programme.

Family law. The revised Family Code 2000 improved the position of married women. Among the changes are (a) women have the right to own and administer common marital property; (b) a husband could no longer deny his wife permission to work outside the home; (c) courts have more rights in settling disputes regarding divorce and inheritance at the cost of traditional arbitrators; and (d) the age of marriage was increased from 15 to 18 years (Hallward-Driemeier and Gajigo, 2015).

The reform of the Family Code is associated with a growth in women’s work outside the home and in higher wages, through the implementation of new rules on paid employment (Hallward-Driemeier and Gajigo, 2015; O’Sullivan, 2016). This has supported the uptake of new jobs in the agricultural and manufacturing sectors, especially by poor and female/youth-headed households. Off-farm income is somewhat more important to female-headed households (17% of total income) than to male-headed households (15% of total income). It remains uncertain, however, whether women are pushed into non-farm activities in order to earn an income, largely because they have less access to resources or own land, which is typically held by older and male-headed households (Bachewe et al., 2016; Snyder and Cullen, 2014), or whether they are pulled into non-farm activities as a result of higher wages (Schmidt and Bekele, 2016).

Land titlingUnder the 1995 constitution, land remains vested in the state, cannot be sold, and the state retains the right to redistribute it. In 1998, a land-titling programme was initiated with the aim of increasing tenure security, ensuring more sustainable use of land resources, reducing conflicts and strengthening women’s rights to land (Alemu, 2015). So far it has registered most

117

rural land in the four most populous regions of Ethiopia: Oromia, SNNPR, Amhara and Tigray. Farmers have use rights, are allowed to lease land (generally up to 50% of the holding) on a short-term basis, can use the land as a collateral, can hire labourers to cultivate the land and have the right to bequeath land within the family, although with some restrictions. Inheritance is allowed only if the family member has no other livelihood support (Deininger et al., 2008; UN Habitat, 2008). Land administration has been decentralised to regions that have enacted their own land proclamations.

Women formally enjoy equal treatment regarding the use, transfer, and control over land and in inheritance of property. The constitution prohibits customary practices that discriminate against women (Girma and Giovarelli, 2013; Gebeyehu, 2014).

Landholders granted use rights were provided with a Book of Holding containing an official certificate. The certificate contains a description of the property, an estimated plot size and the names of the adjacent plot holders. Amhara, SNNPR and Oromia Regions required joint certificates of agricultural landholdings in the names of both husband and wife; in Amhara and SNNPR the photographs of both spouses have to be included (Bezabih and Holden, 2010). Superstitions about being photographed have deterred women from having their photo taken and included in the certificate: a possible explanation for the high number of certificates issued in Oromia exclusively in the name solely of the husband (Deininger et al., 2007).

Where the wife’s name or photo is not included in the certificate, she risks losing the right to claim half the land in the event of divorce or widowhood (Mekonnen and Asrese, 2014).

For women entering into marriage without their own land – property inherited by a spouse prior to marriage is considered personal property, but as traditionally women take up residence in their husband’s community, women rarely inherit land from their own family – their right to land is contingent on the type of agreement entered into on marriage. Although the formal legal system recognises equal rights and prohibits discrimination against women, discriminatory application of laws, enforcement challenges,

high female illiteracy rates and socio-economic constraints often prevent women from exercising their rights (Girma and Giovarelli, 2013). Women’s knowledge and participation in the land-registration process was also found to be generally lower than men’s (Kumar and Quisumbing, 2014) and even in situations where women knew about their rights to land, elders and officials are often reluctant to comply with them (Bevan et al., 2013).

Nevertheless, land certification was seen to have positively influenced investments and land productivity, increased land rental markets and increased women’s participation in land-market activity and even improved child nutrition (Persha et al. 2017). A second phase of the land registration, moving on from just describing land holding and issuing a Book of Holding, to mapping boundaries and recording them in a computerised system resulted in small increases in credit access, moderate impacts on tenure security – that is, increased likelihood that the household believes it can bequeath land – and increased involvement of women in land-related decisions and land ownership – the likelihood that women hold land in their name increased and wives increasingly decided which crops to grow on land in their possession (Persha et al. 2017).

Even though the law provides equal rights to land to men and women, only 3% of all landholders in Ethiopia are young women (Bezu and Holden, 2014). Future prospects for young women to own their own land are not much better: three-quarters of household heads interviewed in one study do not intend to bequeath any land to their daughters and they expect them to obtain land through their husband. This unwillingness contrasts with daughters’ expectations: 41% of young women expect to inherit land from their parents (Bezu and Holden, 2014).

Gender-based violence and harmful practicesGender-based violence (GBV), usually violence against women (VAW) and girls, such as rape, abduction, domestic violence and harmful traditional practices associated with long-term consequences such as FGM/C, are all punishable under the revised penal code (UN Women, 2014). Male violence against women is, however,

118

is widespread across Ethiopia. A systematic review of studies on the topic published between 2000 and 2014 showed that more than 50% of all women experienced physical, sexual or psychological violence from their husband or intimate partner. Almost 75% of women experienced repeated physical assault or battery and close to 40% reported verbal assault, deprivation of freedom to go out, or withholding of money or other family support. Furthermore, more than 25% of women experienced emotional violence and more than half were partially or completely restricted in what they could do without their husband’s permission. Only a minority of women felt sufficiently empowered to report spousal violence to the relevant authorities (Semahegn and Mengiste, 2015).

High levels of acceptance of spousal violence, though declining since 2000, is still widespread despite the revised Family Code and revisions to the criminal law protecting the rights of women and children and promoting gender equality (Figure 77). Factors identified for perpetrating and accepting violence came from the immediate social context such as economic inequality between men and women, levels of female mobility and autonomy and attitudes toward gender roles (WHO, 2005). The degree of social acceptance of violence in society makes eradication hard , while women’s high levels of acceptance of spousal violence illustrates women’s perception of their lower social status and may prevent them from seeking help and justice for themselves and their children (Semahegn and Mengiste, 2015).

Early marriage and early childbearingThe practice of child marriage is still widespread in many regions of Ethiopia for reasons ranging from economic pressure, parents’ wish to strengthen family ties, to ensuring girls’ virginity before marriage (UN Women, 2014; Boyden et al., 2013; Jones et al., 2016). In 2000, the year the Family Code was revised and the marriage age was raised from 15 to 18, 14.4% of all young women aged 15–19 were married at the age of 15. In 2005, this was still 12.7% and in 2011, 8% of girls were married by the age of 15. The median age at first marriage for women in 2011 was 17.1, almost a year younger than

the legal age, whereas for men it was 23.1 (Boyden et al., 2013).

Despite these high levels, child marriage has decreased considerably over the past few decades: in 2011 almost 40% of women aged 45–49 reported that they were married at the age of 15 (CSA and ICF, 2012).

Early marriage results in early pregnancy and, consequently, high fertility. Whereas 12% of women aged 45–49 reported in 2011 that they had their first child at 15, this dropped to 1% for women aged 15–19 in 2011 (CSA and ICF, 2012). This decline in early childbearing is not only positive for young mothers and their health, education and economic activity, but is also crucial for decline of the fertility rate.

Whether a result of the revised family law or a general change in attitudes and the availability of education and employment outside the home, more girls expect to choose their own husband and marriage arranged by parents is in decline. Later marriage for girls is also seen increasingly as beneficial to reduce poverty, power and age imbalances between women and men and to reduce the economic burden of having large numbers of children (Boyden et al. 2013). Increased school attendance and new opportunities for remunerative employment have also contributed to giving girls more confidence and economic freedom, which might have enhanced their decision-making power,

Figure 77 Women’s attitudes towards wife beating in rural areas, Ethiopia, 2000–2011

Source: DHS

0

20

40

60

80

100

Burn

s th

e fo

od

Argu

es w

ith h

im

Goes

out

with

out

telli

ng h

im

Negl

ects

the

chi

ldre

n

Refu

ses

to h

ave

sexu

al in

terc

ours

e

Perc

enta

ge w

ho a

gree

w

ith a

t lea

st o

ne

2000 2005 2011

119

including resisting early marriage and FGM/C (Boyden et al., 2013).

Female genital cuttingFGM/C is still widespread in Ethiopia even though the practice is illegal. Ethiopia is among the top countries still observing the practice, and at least 23.8 million girls and women had undergone FGM/C (Unicef, 2013). In 2016, 65% of all women aged 15–49 reported having undergone the procedure, down from 80% in 2000 (CSA, ORC Macro, 2001; CSA and ICF, 2016). Especially positive is the reduction of FGM/C among women between 15 and 19 years: while in 2000 still 70% of young women had undergone the procedure, in 2016 this was down to 47%; a considerable achievement given that FGM/C is deeply embedded in social norms.55

There seems to be a slight decrease in the rate of decline for the period 2005–2016 compared to 2000–2005, though whether this is a sign of communities rejecting state-imposed rules and counter-reactions observed in some parts of the country that have seen active government and non-government campaigns (Boyden et al., 2013) needs further investigation. There are also considerable regional differences with some regions – Tigray, Oromiya, SNNPR, Addis Abeba and Dire Dawa – seeing reductions of 13% to 18% in the prevalence of FGM/C between 2005 and 2016, whereas regions such as Afar, Somali, or Gambella have seen no change or even increased prevalence.

Some have explained the drop in FGM/C prevalence by increasing secularisation, widespread campaigning against FGM/C, the threat of fines, and that younger women are afraid of admitting that they had undergone the procedure because of the publicity of the law that criminalises it (Yoder and Wang 2013).

The government has a National Committee on Traditional Practices in Ethiopia (NCTPE) with Regional Branch Offices with the objective to eradicate all harmful traditional practices, including FGM/C. It

55 The prevalence in the age cohort of those 20-24 in 2005 should be about the same as the prevalence of the cohort aged 15-19 in 2000. The difference – 71% versus 73% and similar differences between other age cohorts – is normal and to be expected coming from two different samples (Yoder and Wang, 2013).

mounts awareness-raising campaigns and the production and dissemination of teaching materials for primary and secondary schools. In addition, the campaign had strong support and from Christian and Muslim leaders and NGOs (Boyden et al.2013). Concentrating on the negative health impacts of FGM/C in awareness-raising campaigns, rather than presenting the practice as a human rights violation might have been effective in the short term, but might not have contributed significantly in changing the underlying gender norms.

5.2.6 Conclusions for EthiopiaEthiopia has seen considerable economic growth and structural transformation since the early 1990s. Improved macroeconomic policy has led to more economic stability, public investment in roads and other physical infrastructure has encouraged investment, as have shifts in economic policy that have allowed more scope for private enterprise.

Structural change has been limited, however, with most of the change coming from growth in construction and services. The share of manufacturing remains small.

A concerted effort to boost agricultural production and raise productivity, through construction of rural roads, provision of agricultural extension, and supply of inputs and finance through cooperatives, has paid off.

Table 4 National FGM/C prevalence data disaggregated by age cohorts 2000–2016

Age cohort

Year

15–1

9

20–2

4

25–2

9

30–3

4

35–3

9

40–4

4

45–4

9

2000 70.7 78.3 86.1 86.1 83.6 85.8 86.8

2005 62.1 73.0 77.6 78.0 81.2 81.6 80.8

2016 47.1 58.6 67.6 75.8 75.3

Source: DHS, 2000; 2006; 2016

120

Agricultural growth since the mid-1990s has been relatively rapid. That has meant that deep poverty, much of it rural, has been alleviated. A combination of agricultural growth and social protection has meant that harvest failures no longer threaten outright famine, as happened in the 1970s and 1980s.

The RNFE has not yet developed to the degree that might be expected from agricultural growth, limiting the ability of rural households to diversify their incomes away from farming. The demand for off-farm work for rural women has thus been limited.

Rural women have gained from development, largely in material dimensions. Agricultural growth has reduced deep rural poverty, benefiting many rural women. That said, women farmers face notable disadvantages compared to men, in access to land, water, technical information, finance, inputs and labour. Hence when women manage farms, they tend to produce less than men can.

A long and important agenda of remedying the biases against women as farmers can thus be compiled. To that can be added that finding ways to stimulate the RNFE would benefit women as producers; as would measures to aid them when migrating – better information, protection from abuse, and of rights as citizens wherever they may be in the country.

Women’s considerable workload as carers may have eased a little, owing to falling fertility and fewer children, improved health, and in some areas closer access to safe water. The gains, however, have been limited to date.

Some of the most striking gains for rural women and girls can be seen in their human development. Much has been achieved in promoting better access to services such as primary education, primary health care, water and sanitation, and agricultural extension, from which women and girls have benefited greatly. Children’s health has improved significantly, life expectancy has lengthened, and many more girls go to primary school than a generation ago. Family planning services increasingly allow rural women to control their fertility.

Much, however, remains to be done. This is particularly so for education, where still only a handful of rural girls – 6% in 2011 – go to

secondary school. That is mainly for lack of secondary schools within ready access of most rural households: rural boys have similar levels of secondary schooling. Even so, rural girls are often not encouraged to go to secondary school, since adolescent girls are expected to work in the home and the farm. Early marriage has also been the norm.

Trends in rural women’s empowerment are mixed. While more women are economically active both on and off farm, more girls attend school than ever before, and health outcomes have improved considerably, gendered norms and perceptions regarding women’s roles in the household and in society have deep roots and appear to have changed little.

Patriarchal structures are still very strong, though women and girls are increasingly aware of their rights. More women and girls are claiming their right to make decisions regarding the household, their own life and future, for example on early marriage or staying longer in education; but they still face considerable opposition from parents, husbands, elders, religious leaders and officials. Overall, social and economic barriers remain to women’s self-determination and economic status, despite recent legal changes and improvements in services that have enhanced wellbeing.

Many initiatives by government and its development partners work towards changing norms, attitudes and behaviors, but these need to be supported and sustained. This applies particularly to partners who work on raising women’s awareness of their legal rights and improving legal services, capacities and infrastructure that are open to poor and rural women.

Revisions to the family code and a wide-ranging land-titling programme have the potential to benefit women and girls. In principle, the right building blocks are in place that are required for women and girls to lead an empowered economic, social and political life. There are, however, problems in implementation. These include continued and deep-rooted patriarchal norms and attitudes, including among frontline government staff, stalling the implementation of laws that prohibit discrimination and practices that harm women and girls, especially in rural areas.

121

Moreover, too many development initiatives that require public participation can overburden already time-constrained people, especially women. Because of the multiple roles women play in both the productive and reproductive sphere, they lack time to participate in village-level committees, which increases the likelihood that decisions take no account of women’s needs and interests.

Lack of movement toward increased, broad-based political participation, which could cause violence or civil unrest, could undo much of the progress made to date. A restrictive Charities and Societies Proclamation prohibits national NGOs and foreign NGOs that receive more than 10% of external funding from engaging in advocacy in the fields of human rights, women’s rights, children’s rights, conflict resolution or democratic governance. Organisations that worked previously on human and women’s rights have significantly scaled down their operations, including the Ethiopian Women Lawyers Association, which was instrumental in the family law reform process (Amnesty International, 2012).

5.3 Tanzania

After a long period of economic stagnation caused primarily by poor policy choices, Tanzania’s economy is growing, and a nascent structural transformation is occurring. Several factors account for this: more than a decade of macroeconomic stability and financial policies supporting inclusion; a growth process which includes development and transformation of the agricultural and rural sector; and increased efficiency of public spending on growth-enhancing and poverty-reducing programmes and investments. Underlying all of these factors is the cessation, starting in the 1980s but accelerating in the 1990s, of a set of policies which fundamentally retarded Tanzania’s economy and impoverished both rural and urban areas.

Broad-based growth has meant that the structure of employment is gradually transforming toward more productive non-farm employment, and urbanisation is increasing, including secondary cities where 27% of Tanzania’s population now resides. The share

of the population living in poverty is falling. The workforce is better educated and healthier. Women in the workforce have benefited from these changes. Rural women and girls have many more opportunities than their mothers and grandmothers did, and are keen to take them. In many parts of the country, they can and they do.

Tanzania has far to go economically, and some storm clouds are on the horizon. National savings are still low, so fiscal space is limited. High fertility, especially in rural areas, continues to hinder growth in per person income, as well as employment and household transformation. It also burdens an education system struggling to improve quality and expand access to secondary and higher education. Household chores, including caring for young children, continue to burden rural women.

Tanzania’s prospects depend as much on transformation of the non-agricultural sectors as on continued agricultural development. Growth in the agricultural sector depends on demand for food and commodities from households and processors engaged in non-agricultural sectors. The pull of higher earnings in the non-farm sector needs to continue to offer opportunities to the growing labour force so that the share of employment in the agricultural sector can continue to fall.

At the same time, Tanzania cannot neglect the agricultural sector and rural areas. Public investments in market-enabling infrastructure are still needed, as well as support for agricultural exports. Political support for smaller families, combined with accessible and affordable reproductive health programmes using community health workers (CHWs) have been successful in lowering fertility in some other east African countries. In rural schools, a stronger focus on quality and learning would help raise human capital among those about to enter the labour force, as well as those just starting school.

5.3.1 Economic growth and structural transformation in TanzaniaTanzania is experiencing a nascent structural transformation, which is helping to sustain its current growth spurt (Figure 78). Since 2000, Tanzania has been one of the fastest-growing economies in Africa. During this period, the

122

agricultural sector just about kept pace with other sectors, and, owing to rapid growth of the mining and construction sector, the share of industry in GDP rose (Figure 79). The construction and the services sectors generated many new jobs, allowing the share of employment in agriculture to fall to 67%.

Aggregate labour productivity is rising, in part because of increases in within-sector productivity, but mostly because of structural change – labour moving from low-productivity to higher-productivity sectors.

This output and employment transformation has been reflected in increased urbanisation: the urban share of the population is up to 31%.

Tanzania’s structural change transformation is skewed towards non-tradable sectors, a common pattern in sub-Saharan Africa. Foreign aid has allowed imports to rise as a share of GDP, while exports have remained roughly constant – albeit with diversification owing to growth in mining and tourism exports; unprocessed commodities no longer account for the majority of Tanzania’s exports. Manufacturing exports have grown as well, mostly to regional trading partners.

As a share of GDP, manufacturing has remained roughly stable, reflecting high growth from a very low base. It has become more labour-intensive – and informal – accounting for a slightly higher share of employment.

The main driver of structural change over the last 15–20 years in Tanzania has been

dramatically improved macroeconomic management, abandonment of the socialist model of development, and a deregulation of key parts of the economy (including agricultural trade). After finally conceding the failures of the state-owned and operated model of development, Tanzania began a reform programme in the late 1990s. State-owned enterprises and banks were privatised (or closed), and private investment was encouraged. Donors gradually supported the programme, increasing concessional aid and grants (Figure 80).

As a result, public and private investment have risen, and the financial sector has deepened as credit to the public sector has risen as a share of GDP. Private banks, microfinance, and mobile money are all thriving. Public spending has focused mainly on infrastructure development and delivering social services, two key drivers of increased labour productivity. FDI has also grown as international confidence in Tanzania’s macroeconomic management and economy improved.

As noted, trade and increased integration in the global economy have not yet played a strong role in transformation. This must change as Tanzania needs new technology and to be competitive in global markets. Constraints to the development of a more open and diversified economy include:

• deficits in transport infrastructure, including poor port performance and high transport costs

• energy shortages

Figure 78 Economic growth, Tanzania, 1988–2015

Source: World Development Indicators

0%

2%

4%

6%

8%

10%

400

450

500

550

600

650

700

750

800

850

Annual Change, GDPGD

P pe

r ca

p.,

US$

cons

tant

201

0

GDP annual change GDP per cap

123

• continued weaknesses in microeconomic policy, including policy inconsistency

• slow creation of a new modern sector firm, leaving most people still working in household farms and firms

• land policies, which make urban and industrial development expensive and difficult

• weak domestic resource mobilisation, which limits the amount of public investment that can be undertaken.

5.3.2 Population changeTanzania’s population has grown rapidly in recent decades: growth rates have varied between 2.5% and 3.5% a year, with no clear sign of slowing down (Figure 81).

Tanzania is urbanizing, although most of the population remains rural: the urban share of population rose from 17% in 1985 to 31% in 2015.

High fertility (see below) is probably reducing public and private savings, as resources are needed to feed, educate, and house a rapidly growing young population. High growth of the labour force has both limited employment transformation and wage growth in enterprises.

56 Some efforts were made to establish estates and plantations, such as those for sisal in Tanga; but these produced only isolated, small enclaves of more capitalised farming. Unlike Kenya, little attempt was made to stake out substantial areas for white settlers.

5.3.3 Agricultural Transformation in Tanzania

Background: geography and policyGeography and policy have proved critical to agricultural development in Tanzania since independence in 1961. With almost 1 million km2, Tanzania is a large country – 30th in the world, 11th in Africa, with a population which in 1961 was just 10 million, and is now 53 million. Natural potential for agriculture, however, has been difficult to realise owing to the costs of long-distance transport in a country that until recently has been overwhelmingly rural with few cities of any size. Except for those relatively few farmers living close to cities and ports, their remoteness from urban centers has meant that they have faced low effective demand at the farm gate. That same distance also meant that new technology embedded in better seeds, fertiliser, irrigation equipment, veterinary medicine and so on, was either unavailable locally or only at a high price.

Ever since the colonial invasion of the late nineteenth century, governments have repeatedly tried to engage smallholders56 to grow larger surpluses. Agricultural officers would introduce new crops, demonstrate more intensive

Figure 79 Share of GDP by sector, Tanzania, 1990–2015

Source: World Development Indicators

0

20

40

60

Agriculture, value added (% of GDP)

Industry, value added (% of GDP)

Services, etc., value added (% of GDP)

Figure 80 Improvement in Pillar Indices, Tanzania, 2012–2015

Source: World Economic Forum, Global Competitiveness Report

6% 5% 4%

14%5%

-13%

1%

-15%-10%

-5%0%5%

10%15%

prop

erty

rig

hts

publ

ic s

ecto

r pe

rform

ance

corp

orat

e et

hics

qual

ity o

f roa

ds

dom

estic

com

petit

ion

fore

ign

com

petit

ion

trade

tar

rifs

124

production techniques, but often to little effect. Low prices at the farm gate resulting from long distances to cities and ports were unattractive. Subsistence farming dominated, with farmers generally using land extensively with low yields per unit area.

The abundant land, however, varies in quality. Some 30% of the land is arid, unsuitable for farming unless irrigated, and another 30% is semi-arid, of low potential for crops. That still leaves 40% with medium to high potential. Commercial agriculture has long flourished in some of the latter areas. For example, the slopes of Mount Kilimanjaro and Mount Meru contain rich volcanic soils and receive sufficient and quite reliable rains. Farming in these areas has long been intensive, with valuable cash crops such as coffee. Hence unevenness is a feature of agricultural development.

The other critical factor has been policy.57 From the mid-1960s to the late 1980s, Tanzania pursued radical policies to develop the country that for the most part failed. In the agricultural sector, the state organised input supply and marketing through parastatals, the largest being the National Milling Corporation, that had taken over from the existing farming cooperatives by 1976. Pan-territorial prices

57 Main sources: Delgado and Minot (2000); Dercon (1994); Ellis (1983); Hyden (1980); Putterman (1995).

were paid for key crops: maize, cotton, rice, wheat, oilseeds, and cashew.

These measures were accompanied, from 1973, by villagisation (ujamaa vijijini). Rural households, many of whom lived in dispersed across the landscape, were resettled into concentrated villages (vijijini) to make it easier for the state to provide education, health and water; as well to promote more collective forms of production (the ujamaa).

Agriculture suffered badly from such policies. The parastatals ran up high costs through inefficiencies and replication of warehouses and transport fleets specific to crops. These costs were discounted from the prices paid to the farmers. To make matters worse, over-valuation of the Tanzanian shilling meant that export crop prices were artificially low, while imported food was cheap. Farmers had little incentive to invest and innovate. In some areas, parallel channels, technically illegal, emerged in which traders offered better prices to farmers: but only in areas close to roads and cities.

Villagisation also hindered farmers. Some had to trek for miles to reach their fields next to which they once lived. The communal plots (shambas]) were unpopular, farmers worked on them reluctantly, and produced little: there was

Figure 81 Population change, Tanzania, 1985–2015

Source: World Development Indicators

2.0%2.2%2.4%2.6%2.8%3.0%3.2%3.4%

0

10

20

30

40

50

60

Mill

ions

Population Growth population, % annual

Figure 82 Net rate of assistance to agriculture, Tanzania, 1976–2004 Population change, Tanzania, 1985–2015

Source: Compiled from Estimates of Distortions to Agricultural Incentives, 1955 to 2007, Anderson and Valenzuela, World Bank

-0.7

-0.6

-0.5

-0.4

-0.3

-0.2

-0.1

0

125

simply no point in collective production, while transaction costs mounted.

Agricultural growth thus slowed markedly from the mid-1970s (Figure 82). The failure of this agricultural development model in a country that was overwhelmingly dependent on its farm economy led to economic crisis in the 1980s. A near-bankrupt state – the crop parastatals ran up huge debts with the Central Bank – had to accept IMF recommendations from 1984 onwards.58 The shilling was devalued, farmer prices were raised, price controls removed, private and cooperative trading reintroduced, and the parastatals were either disbanded or were much diminished in their functions.

Since the early 1990s, however, agricultural policy has been less turbulent and more benign. By the early 1980s, the net rate of assistance to agriculture (NRA) had fallen to less than minus 0.63 – in effect a more than 60% tax on farming: by 1993 it was much less, at -0.06. (Figure 82). Although this rate has subsequently fluctuated, the latest estimate of the NRA indicates the effective tax was less than 20%.

Agricultural growth and performanceThe long-term record of agricultural growth shows three periods (Figure 83): from the early 1960s to late 1970s when annual growth rates were typically in the range 3–5% a year; the slowdown noted between the late 1970s and the mid-1990s when annual growth was rarely above 2%; and the subsequent revival of growth where rates of 5–7% a year have been seen in almost half of the years. Consequently, output that had risen by less than 2.5 times from the 1960s to the mid-1990s, almost doubled within the next dozen years.

A good part of the growth of agricultural output seen from the 1960s to the 1990s can be attributed to simple additions to the labour force: labour productivity even fell from US$470 a year per worker in the mid-1960s to US$390 in 1999 (in constant 2004/06 US dollars) (Figure 82). Subsequently labour productivity has risen

58 Reforms came in stages. 1986 saw the Economic Recovery Programmeme that included further devaluation, government spending cuts, import liberalisation, ending of price controls, and setting positive real interest rates. For agriculture, private trading was legalised from 1986 onwards. By the early 1990s a fertiliser subsidy had been phased out. But macroeconomic management remained weak between the start of the programmes in 1986 and the late 1990s, with multiple currency crises, and Privatisation of loss-making parastatal enterprises was slow.

to US$580. Arable land was also added until the mid-1980s, but then declined a little before increasing again from the early 2000s onwards. Land productivity in the 2010s had increased to roughly double the levels seen in the early 1960s.

The rise in farm output seen since the late 1990s can only partly be accounted for by additions of labour and land. Indeed, total factor productivity in agriculture rose by 17% between 1997 and 2013 – compared to just a 3% rise from 1962 to 1997. Hence agriculture has been intensified and become productive since the reform programme took hold.

By crops, some of the largest increases registered since the late 1990s include oilseeds (groundnut, sesame, sunflower), banana, maize, rice, sugar cane and sweet potato. Of livestock products, milk production stands out. In all these products, output has risen by a yearly average of more than 6% a year.

Three factors probably account for the apparent sharp acceleration in agricultural growth since the late 1990s. One is the ending of price repression, although that was achieved almost a decade before the acceleration.

A second factor is the combination of growing urban areas, rising urban incomes and improved roads resulting in increasing number of farmers Realising higher effective demand at the farm gate for surpluses. Increased opportunities and greater incentives are repeated themes in local-level accounts of change in rural Tanzania (see Box 9 below). Urban demand, for example, probably explains why oilseeds have expanded as strongly as they have.

A third element – difficult to confirm other than by reference to micro-studies – may be that by the late 1990s, the dislocations of changing policy for marketing, input supply, technical advice and finance had subsided; with time for private traders, contract farming schemes, and other similar arrangements to provide smallholders with better access to inputs, technical advice and finance.

126

Agricultural trade balanceUntil the 1990s, the bulk of Tanzania’s exports came from agriculture. In the 1960s and 1970s, a large surplus was realised on agricultural trade (Figure 84). From the mid-1970s to the early 1990s, however, exports fell by two-thirds in value, almost eliminating the export surplus in the process. Subsequently exports have revived, growing more than three times during the 2000s. Agricultural imports, however, have also risen equally rapidly so that only a small surplus has been achieved in most recent years.

Assessing Tanzania’s agricultural development The history of Tanzania’s agriculture since independence includes almost 20 lost years, between the late 1970s and 1990s, under the baleful effects of misguided policies and their reform. Had agricultural output increased between 1979 and 1998 at 4% a year – as it had been doing in the five years prior to 1979 – then current output would be over 50% larger than it is. The costs of economic failure can hardly be understated.

For the more recent period from the late 1990s and agricultural recovery, how well has agricultural development performed? Taking

59 Although not necessarily demand. Since the early 1990s, some large increases have been seen for imports of foods such as wheat, palm oil, sugar in various forms. Most of this will have been used in food manufacturing to produce food for urban populations; much of it convenience food, often highly processed food with high levels of sugar, salt, fat and energy per unit weight (namely, ‘junk food’).

the four roles set out by Johnston and Mellor in 1961, then:

• Agricultural growth has been greater than population growth, with strong growth of most cereals. Agriculture has been able to meet domestic needs.59

Figure 83 Agricultural production, Tanzania, 1962–2011

Source: Compiled from FAOSTAT data, Three-year moving averages taken to minimise the effect of yearly weather varia-tions. G=Giga=109

(2.00)

-

2.00

4.00

6.00

8.00

10.00

1

2

3

4

5

6

7

8

9

Annual Growth, 3YM

A, %Gr

oss

Valu

e, U

S$ In

t 04/

06, G

, 3YM

A

Annual Growth, 3YMA Agri Gross Prod Value, US$ Int, G, 04/06, 3YMA

Figure 84 Agricultural trade, Tanzania, 1961–2013

Source: Compiled from FAOSTAT data. Values in US$ deflated to constant 2010 values using US GDP deflator

- 200 400 600 800

1,000 1,200 1,400 1,600

Valu

e, U

S$M

, 201

0 co

nsta

nt

Exports Imports

127

• In most years since 1998, a surplus on agricultural trade has been realised, albeit quite a small one.

• Farm incomes have risen so that demand for other goods and services by farm households has been growing.

• Given rising labour productivity, agriculture has been able to release labour to other sectors. Little, however, is known about capital transfers out of agriculture.

The arguments so far have had to be based on data that are not that accurate. Moreover, the estimates are at national level. Yet Tanzania shows great variations across its diverse geography. Only studies at village and district level capture this diversity. Some of these studies suggest more investment and innovation in agriculture than is apparent when subsumed in national statistics (Box 9).

These cases show that while across much of Tanzania, agricultural development may be proceeding rather slowly, there are plenty of islands where growth has been more rapid. The key question concerns the scope for expanding the islands and replicating promising experiences.

5.3.4 Household TransformationHousehold sizes in rural Tanzania have increased since the early 1990s: the average size of a rural household in 1991/92 was 4.4 persons, which has risen to 5.1 by 2015/16. Female headship has also become more common, up from 17% of rural households in 1991/92 to 24% by 2015/16. [Statistics from Demographic and Health Surveys for 1991/92 and 2015/16]

The more robust and sustained economic growth and the nascent structural and rural transformations of the last 15 years brought more diverse income portfolios to Tanzanian households and reductions in monetary poverty. Although incomes and welfare appear to have risen faster in urban areas, rural households have benefited as well, primarily through development of the RNFE, but also through diversification into new crops and better transport linkages from rural villages to nearby town and secondary cities. Rural inequality remains low, indicating that growth has been shared widely.

Tracking monetary poverty over time in Tanzania is difficult, owing to both poor quality and inconsistent survey data. The quality of household survey data has improved over the last 10 years, but it is still hard to establish a trend. Measurements using the international poverty line show a clear trend of improvement

Figure 85 Arable land, agricultural labor and productivity, Tanzania, 1961–2011

Source: Compiled from FAOSTAT data, labor data from the GGDC 10 Sector Database [Timmer et al., 2015]. Land productivity is all output divided by arable land: since this includes some livestock from pastures, it overstates yields per hectare.

300

350

400

450

500

550

600

650

700

750

4

6

8

10

12

14

Out

put

per

ha/w

orke

r US

$ co

nsta

nt 0

4/06

Arab

le, M

ha

& Em

ploy

men

t M

Arable land Agricultural Labour Land Productivity Labour Productivity

128

since 2000, as rural poverty has fallen from 88% in 2000 to 48% in 2012. Estimates using the national poverty line show a slower trend.60 One reason is that these estimates, based on measures of consumption, exclude expenditures on durable assets, education and health care, three categories which are highly income elastic. Multi-dimensional poverty, calculated using education and health outcomes plus household assets, shows only a slight decline in rural areas between 2008 and 2010.61 This reflects the ‘baked in’ characteristic of this index – variables such as share of adults in the household with primary education do not change even as the household becomes richer; in this case, the share

60 World Bank, “Tanzania Poverty Assessment” (n.d.)

61 The multidimensional poverty index is not available before this date.

of households judged to be deprived only falls as more households are formed by younger, more educated members.

Household employment transformation enabled much of the poverty reduction, especially in rural areas. Although farming is still the main activity of most rural people, non-farm activities have grown in importance. Nationally, only 65% of the population now state that farming is their main activity; it accounts for about 34% of household income. (de Magalhaes and Santaeulalia-LLopis 2015) The number of households reporting some non-farm wage income has grown nationally and in rural areas: by 2014 a significant portion of rural households

Box 9 Insights into agricultural growth from village and district studies

Village and district studies undertaken since the 1980s frequently describe processes of intensification of farming and of diversification into the rural non-farm economy – given the right circumstances. So what do these accounts have in common?

• In all cases, farmers had access to land of at least medium potential in quality of soils and rainfall, and sometimes with the possibility of irrigation.

• The stimulus of a market opportunity, usually in the domestic market. Tanzania may have urbanised slowly, while urban incomes have also increased only modestly: yet that has still created enough demand to pay attractive prices for higher-value produce – fruit, vegetables, oilseeds, dairy.

• A new road, or a better maintained road, that cuts transport costs and time; and, in some cases, the intermediation of a trader or group of traders who spotted the opportunity to produce for a distant market and were able to convince farmers to commit to production.

Few of these cases involve direct government intervention to stimulate production. Often, it is the farmers and traders who find ways to produce surpluses, once they realise that the surplus can be sold at a profit.

The most dramatic examples of local initiative concern irrigation, where groups of farmers have proved able to irrigate tens and hundreds of hectares: enterprises that require capital and cooperation to operate the schemes.

Source: Boesen and Ravnborg (1992); Ponte (2001); Rasmussen (1986); Grove (1993); Andersson (1996); Bantje (1986); Mutabazi et al. (2013); (Mdoe and Wiggins 1997); Molony (2008); Woodhouse et al. (2016)

Note: Blogs from Long-term Livelihood Change in Tanzania that report initial insights from planned resurveys of villages first studies in the 1980s and 1990s, with nine cases so far recorded, at http://livelihoodchangeta.wixsite.com/tanzania/blog

129

Figure 86 Employment structure and growth in non-farm activity

Source: Tanzania Labor Force Survey; main activity only

0

5

10

15

20

25

Urban Rural All

Mill

ions

Wage HE Farm

-1%

1%

3%

5%

7%

9%

Urban Rural Total

Annu

al g

row

th r

ate

Wage HE Farm Total

Figure 87 Household sources of income

Source: Tanzania Labor Force Survey

0

20

40

60

80

100

Urban Rural Total

% o

f hou

seho

lds

with

inco

me

sour

ce

Farm HE Wage

0

20

40

60

80

100

Urban Rural All

% o

f hou

seho

lds

with

inco

me

sour

ce

Farm HE Wage

130

had someone engaged in a non-farm activity at least part of the year (Figures 86 and 87).62

In part, this reflects the importance of secondary non-farm activities, to reduce seasonal underemployment, but it also reflects household livelihood diversification. However, the number of employed reporting a secondary activity is smaller in 2014 than in 2006 (Figure 88). It also reflects urbanisation – urban activities are not subject to so many seasonal fluctuations. But even in rural areas, the share of the labour force reporting secondary activities in the RNFE is lower. Agriculture accounts for 56% of rural incomes, a much higher share than in Uganda, where it is only 33%, even though Uganda and Tanzania have about the same total average household income in rural areas. (de Magalhaes and Santaeulalia-LLopis 2015) This high share may reflect the shift of rural cropping patterns described above, away from just growing for subsistence into newer cash crops, irrigation, and livestock, which provide year-round agricultural income-earning activities. Household enterprises (HEs) – household-based rural non-farm businesses that rarely, if ever, employ anyone outside the family – still make up an important part of household livelihoods in both rural and urban areas.

Household welfare has also improved. The provision of public services has allowed rural households to increase human capital.

Qualitative data provide some additional insights into how lives in rural households have changed. They confirm the quantitative data – higher incomes from farm and non-farm activities have driven reductions in poverty and deprivation. They confirm that most households now produce cash crops and make a better living, as demonstrated by increased assets and new products and services available for purchase in the villages.63

62 The growth of farming as a main activity in urban areas reflects a change in sampling frame as rural towns and peri-urban areas was reclassified to the urban strata in the latest survey. It does not reflect a surge in urban farming; see national numbers.

63 Blogs posted by researchers in the study of long-term livelihood change in Tanzania. The study project revisits households and communities in Tanzania that were surveyed by diverse researchers in the 1990s and early 2000s, to build up longitudinal datasets exploring economic change and poverty dynamics. Fifteen sites and 70 villages are included: http://livelihoodchangeta.wixsite.com/tanzania/blog

• Arusha: coffee was replaced by tomatoes, other vegetables, creating better livelihoods.

• Njombe region: pine trees and potatoes are key cash crops, offering year-round income.

• Mlali village: vibrant land market; “tractors now used for land preparation; no sign of this service 20 years ago”.

• Ikuwala village: “tomato market has grown and prices have increased; motor cycles are ubiquitous”.

• Mtowisa: many pigs, small livestock, which was unusual in 2000; new cash crop – oil seeds; petrol pumps used to draw water.

• Songea rural district: hybrid maize seeds widely available; new crops include pigeon peas, soybeans, oil seeds; cashew revival starting; ‘transport much improved – year around bus service’.

• Goima: now connected to national electricity grid.

• Everywhere: many new shops in villages; major improvements in housing quality.

• Gitting: ‘Many women report taking microfinance and starting own business’.

Drivers of change reported in the qualitative data include high land ownership, which provides security, as well as public expenditure on rural roads, education, health, and water supply. TASAF, the social action fund, originally funded village infrastructure but has now been converted into a rural conditional cash transfer (CCT), providing support for the poorest rural households. Some improvements in living standards can be attributed to this. Village banking (SACCOs and savings groups), which have existed for 20–30 years in some areas, have now combined with microfinance and mobile money to bring rudimentary financial services to rural households. This has also enabled urban–rural money transfers. Information also flows more easily

131

between cities, towns, and villages through the expansion of the mobile phone network.

Rural households, which tend to be younger owing to higher fertility, find that increased claims on household resources have come with increased income. With more educational infrastructure and the perception among rural households that education is the current route to better livelihoods comes pressure for spending on school fees, uniforms, and other education-related costs. Drugs for treatment of illnesses still must be purchased by households at private pharmacies since public clinics often run out. These household expenditures compete with opportunities to invest in higher-value agriculture or the development of non-farm activities.

In sum, agricultural transformation has been the main driver of improvements in welfare for rural households. Some good initial conditions, such as rights to land ownership and in some places good soils and access to water for irrigation, have been enabled by better policies and more connectivity. Connectivity has enabled agricultural diversification into new crops for

64 Though anecdotes of wives fleeing violent or oppressive husbands are all too common (see for instance mentions in Smith and Sender (1990), stories of family harmony are hopefully more run-of-the mill, tending not to feature as unusual cases in the literature. That said, the 2015/16 DHS for Tanzania reported some 38% of rural married women aged 15-49 had at some point experienced emotional violence committed by their spouse or intimate partner, while 41% had experienced physical violence and almost 14% sexual violence.

new markets. Progress has been uneven, of course. Areas with less water, and pastoral areas are noticing the effects of climate change in more volatile weather patterns. Not surprisingly, insecurity of income remains a concern for farm households.

5.3.5 Lives of rural women and girls: gendered aspects of household transformation in rural areas

Women as economic participantsA look at the literature to determine the extent to which farming is carried out jointly or separately by men and women in Tanzania reveals – unsurprisingly – a great deal of variety. Few publications, though, really attempt to answer this question from the perspective of a cooperative or non-cooperative64 family unit – focusing instead on gendered tasks, products, assets and the like across whole villages or cultural systems. The reality of family farming is more nuanced than ‘men control cash crops, women control food’: it depends on all sorts of

Figure 88 Number of employed people with secondary activity by type

Source: Tanzania Labor Force Survey

0%

20%

40%

60%

80%

100%

Urban Rural All

Wage HE Farm No secondary activity

0%

20%

40%

60%

80%

100%

Urban Rural All

Wage HE Farm No secondary activity

132

factors including relative wealth and life cycle for the men and women involved. In areas with more water and some irrigation, farming tends to be a shared activity. In pastoral households, there tends to be a stronger gendered division of labour, with men and boys handling the grazing/watering of the livestock and women tending to the household gardens and food plots. Women tend to be responsible for milking cows, and sometimes for taking the milk to collection points and therefore exerting some control over the income.

Contrary to reports of the feminisation of agriculture, the division of farm labour is reported to be fairly even in the sense that in many areas women and men spend about the same time farming (Mbilinyi, 1994). Women do work far longer days than men, however, owing to their labour in processing and marketing agricultural produce and their heavier burden of domestic and unpaid care work. Several farm tasks are also done by both sexes (FAO Links 2003; Smith 2015). In harvesting and processing maize, for instance, both tend to be involved in de-sheathing, transporting, piling, drying, packaging, and storing cobs (Rugumamu 2009). Nonetheless, even when working on the same tasks, men and women may use different technologies or tools. For instance, in de-sheathing maize, women may be more likely to use knives or their bare hands. In transporting cobs from fields to homes along tracks, most women carry them as headloads, while most men use wheel barrows or hired bicycles (Rugumamu 2009).

Has the nature of agricultural development affected gender relations? Some of the growth sectors of farming perhaps favour women’s participation. Dairying and small-scale vegetable growing would be examples, since they are activities that can be carried out close to the house and can be interspersed with domestic tasks. That, of course, may mean that women carry a heavier workload: a cross-bred dairy cow can drink 80 liters of water a day, and needs to be milked twice a day, a chore which usually falls to women.

In areas where land was customarily allocated to households, women have not traditionally owned land but have obtained access through marriage; then losing it through widowhood or divorce, unless brought into a larger family unit (such as brothers-in-law), creating its own set of

challenges to women’s autonomy. Women own about 19% of all titled land in Tanzania, while holding sizes for women are on average less than half of those of men (from about 0.21 – 0.3 ha, compared to 0.61-0.7 ha for men (Leavens and Anderson 2011). Where customary land rights allocate land to unmarried women, it tends to be in smaller parcels – as clans assume a woman will marry and leave the area. Unmarried women often work as farm labourers because they have no land of their own, but upon marriage they are usually allocated a parcel of land and in some places a cow or cows to milk.

The growth of the RNFE has been a boon to women, as they have started businesses in large numbers. Some operate the business jointly with their husband, but most do business on their own (Kweka and Fox 2011). Labour Force Survey data suggest that for rural women, having a household enterprise is a more recent phenomenon. Some 10% of male owners report having had their business 20 years or more compared with 5% of women, while 44% of men created their business in the last five years compared with 50% of women. Men and women tend to operate different kinds of business: there is occupational segregation by gender in household enterprises (Filmer and Fox 2014).

In rural Tanzania, the most common business among both men and women is trading, as it is nationwide (Kweka and Fox 2011). Men dominate mining and construction, while women dominate accommodation and food services (Figure 89). In Gitting, villagers report:

Women in particular have benefited. Many women report taking microfinance and starting own businesses - not possible in 1991 owing to lack of financial access but also attitudes to women running businesses.

Women are entitled to hold bank accounts and take credit in their own names, although formal banking remains limited to urban areas and for those with collateral and/or a salary. Tanzania has become awash with informal credit of different sorts, and it has been claimed that women have easier access to credit from many organisation due to positive discrimination.

133

For women in the labour force, nationally the main differences in main economic activity between women and men are in urban areas. Women are less likely to have wage employment, and more likely to be running a household enterprise (Figure 90). Part of this reflects the expansion of the construction and transport industries, where women rarely work. In other sectors, such as manufacturing, women’s share of wage employment has grown. In rural areas women are also less likely to be in wage employment, but overall, non-farm wage labour represents only a tiny share of rural employment.

Women as mothers, wives and carersBetter opportunities in rural areas in farm and non-farm activities have not reduced the burden of household chores for women. (Figure 91)

Large families are part of the story. Multi-generational family compounds still help with childcare tasks, with younger women in productive activities and older women looking after children. But starting as early as 10 years of age, girls spend a significant portion of their time on domestic chores, which limits their time in productive activities.

One of the most time-consuming tasks is fetching water and wood. While the number of households with water on premises, such as a well, has risen substantially owing to private investments, the percentage of households having to travel more than 30 minutes to get water has not budged since 2000 (Figure 91). If a household has a bicycle, boys may be sent to fetch water or collect wood. Only 6% of rural household report access to electricity for lighting.

In some richer villages nearer to urban areas, maize-milling machines have become more widely available, saving labour for women. Processed food is also gradually becoming more common in the rural diet.

Women’s human developmentIn education, primary school enrolment is almost universal, and has been so since about 2010 (Figure 92). The gross primary enrolment ratio was over 100 in 2010 while net enrolment

was about just under 80%, reflecting the tradition of starting children in school late, especially in rural areas. By 2012, 65% of the population over age 25 had completed primary school, and this number has increased as younger cohorts enter adulthood.

Secondary enrolment is also increasing as the school network expands, although cost and distance remain major deterrents to enrolment for rural households. By 2015/16 girls were slightly more likely to be enrolled in both primary and secondary school than boys.

In health, under-five mortality in rural areas has fallen considerably since 2000, more than having (Figure 93).

Less progress has been made on reducing maternal mortality: the national rate fell from 118 to 87 per 100,000 between 1996 and 2016.

Some rural areas report improved access to safe water, while in others, no progress has been registered; nationally, the improvement the number of households in rural areas with water within 30 minutes’ walk is marginal, although the share of rural households reporting water on premise has grown, reflecting private investment by richer rural households in wells on their property (Figure 94).

Tanzania has only weak institutional support for family planning. Contraception is free in government facilities but supplies are often intermittent, and it is not actively promoted. There is no consistent message from central government, and many mixed messages given the powerful role of churches and mosques at village level. The Catholic church is immensely powerful, and there is a plethora of new evangelical Pentecostal congregations, and Islam is important in coastal regions Islam. Thus, although age of first marriage/union has been steadily rising in rural areas, rural fertility remains high at over six children. Demand for contraception is low. Only 35% of married women in rural areas report using any kind of contraceptive method, and 18.5% report an unmet demand for contraception according to the 2016 DHS survey. This means that only slightly over 50% of rural married women even

134

Figure 89 Sector of activity of rural household enterprises, 2014

Source: Labor Force Survey

22%

12%

7%44%

6%6%

3%

4%12%

0%

52%

25%

1%6%

Figure 90 Structure of employment, main activity, by gender and area

Source: Tanzania Labor Force Survey

5

0

10

15

Men Women All

Mill

ions

Wage HE Farm

5

10

15

Men Women All

Mill

ions

Wage HE Farm

0.0

1.0

2.0

3.0

4.0

Men Women All

Mill

ions

Wage HE Farm

2

0

4

6

8

Men Women All

Mill

ions

Wage HE Farm

Mining and quarryingManufacturing

Construction Wholesale and retail

Accommodation and food service Transport, storage and communication

Other

Male-owned household enterprise Female-owned household enterprise

135

want to use contraception in Tanzania, despite rising education levels.65 However, as with other aspects of life in Tanzania, fertility varies by region, tribe, and culture. Fertility is highest in Western Tanzania, especially in pastoral areas in the North, and lowest on the coast (Figure 95).

Women as citizens and agents, gender normsAlthough united by one language (Swahili), the organisation of households and the extent to which laws and gendered norms determine the rights and opportunities of rural women and girls in Tanzania is heterogeneous, shaped by historical tribal customs and cultures as well as local conditions. For example, formal family laws set a minimum age of marriage (14 years), as well as procedures for divorce, control of assets including children in case of widowhood, etc. This law confers ownership of children on their father in case of divorce; they can stay with their mother if they are under seven years of age, after which they should be with the father. Sexual activity with a girl under the age of 14 is illegal.

But access to the formal courts is mediated by wealth so that most the population have no access to this system. At village/community level a mixture of cultural and family practice and religious guidance applies, while local government actors may get involved. The village council has a peace and security function and often mediates in domestic and family disputes, seeking peaceful resolution rather than prosecution. Religious leaders often play a significant role particularly in relation to marriage and divorce.

Owing in part to traditions of joint farming activities, Tanzanian women have long had influence over family spending decisions. However, as women have earned more income, qualitative work suggests that they have acquired an even greater say in household decisions; “it is now considered usual for women to play an equal

65 It would not be expected that all women of child-bearing age would want to use contraception all the time, as some may be pregnant, breastfeeding, or trying to get pregnant. However, Tanzania’s numbers are low. In Rwanda, a poorer neighbouring country in which women are less educated, 53% of married women in rural areas reported using contraception in 2014-5.

66 This is common among adolescent boys, who tend to mature more slowly than girls. See (Bertrand et al., 2013).

role”. Men do not want women to stop bringing in money: they recognise that women’s increasing role in earning cash income is a necessity.

Tanzania’s laws and policies ensure equal access to education and health services for men and women, and local norms mostly support this. Indeed, Tanzanian girls are enrolling more often than boys, and reportedly doing better in school. In qualitative studies, villagers complain about boys not applying themselves in secondary school and having a very short time horizon.66

Although Tanzania’s laws forbid most gender-based violence (principally VAW), it remains common. Forty per cent of rural married women have experienced physical violence from their spouse or intimate partner. Support for such abuse appears to waning, however, especially among rural men. In the village of Vitonga, qualitative analysis finds that women joined to convince male elders to stop sanctioning physical violence against women, and it is now much less prevalent. In pastoral areas, where male and female roles are more distinct, VAW is reported to be more prevalent.

Figure 91 Time spent on productive activities and household chores, Tanzania, 2006 and 2014

Source: Labor Force Surveys

- 50

150

350

Female Male

Productive activities Household chores

2006 2014 2006 2014

136

5.4 Conclusions for Tanzania

Similar to Ethiopia, Tanzania has seen renewed and strong economic growth in the new century, including in agriculture. Tanzania’s development differs from that of Ethiopia in two significant respects. One is that rural non-farm economy is larger in Tanzania; the other is that population growth remains rapid in Tanzania, since fertility has fallen only slightly in recent decades.

As producers and participants in the labour force, women have clearly gained from the economic opportunities created by rural transformation in Tanzania. As mothers

managing the household, women’s work in the care economy has not reduced: if anything, it may have increased. Without recourse to modern contraception to limit the number of births, it is hard to see how this workload will diminish, although improvements in rural water supply would help. Increased incomes are allowing the purchase of more processed food, reducing the time needed for cooking.

Girls’ opportunities have improved dramatically, especially given improved access to education. Parents support girls’ education, which they see as a good investment, possibly even a better one than educating boys.

As participants in their household and community, women have gained stature. The age of marriage is gradually rising, and women are increasingly seen as equal partners in the household, and as active participants in community affairs – to the extent that their paid and unpaid activities allow. That said, gender relations are far from equal.

5.5 Conclusions. Insights from Ethiopia and Tanzania

5.5.1 Nascent inclusive transformationsBoth Ethiopia and Tanzania have recovered from the damage of post-independence economic policies that impeded rural transformation. Since the early 1990s, both countries have invested public resources in social and economic

Figure 92 Net primary and secondary attendance, Tanzania, 2004/05 to 2015/16: girls are doing well

Source: Compiled from DHS reports

0

50

100

2004/05 2010 2015/16

Net a

ttend

ance

, %

Female national Female ruralMale national Male rural

Primary

Primary

Primary

Seco

ndary

Seco

ndary

Seco

ndary

Figure 93 Under-five mortality, Tanzania, 1992–2016

Source: Compiled from DHS reports

60

110

160

210

Unde

r-fiv

e m

orta

lity

rate

(p

er 1

,000

live

birt

hs)

National Rural

1992 1997 2002 2007 2012

Figure 94 Access to water, rural area, Tanzania, 1992 to 2016

Source: Compiled from DHS reports

0

20

40

60

% o

f hou

seho

lds

Water less than 30 minutes awayWater on the premises

1992

1996

2000

2004

2008

2012

2016

137

Figure 95 Fertility, by region, Tanzania, 2015/16

Source: DHS Report, 2015/16

infrastructure to support private investment instead of directly in productive enterprises. In both countries, substantial foreign aid in the 2000s financed these investments, similar to the role US government resources played in Thailand. Partly because it did not suffer significant internal conflict since independence, Tanzania is well ahead of Ethiopia in per person income, connectivity, economic diversification, urbanisation, and social indicators.

Economic diversification has taken place in both countries as the share of agriculture in national output has declined and the share of services has risen. Unlike the three countries in the previous chapter, a robust manufacturing sector has not yet developed to drive structural transformation, although the potential exists in both agro-processing for export and agricultural production for a growing urban population. This has meant that when employment and livelihood diversification has taken place, it has primarily been within the household sector rather than from the household sector to the enterprise sector. As a result, the share of the labour force with a full-time wage job as the main activity has been growing slowly.

Rural areas are transforming rapidly. Ethiopia still has many subsistence farmers, while in Tanzania most farmers produce in part for the market, and report higher input use than Ethiopia, another indicator of transformation. An indicator of developments in connectivity

and market development, farmers in Tanzania report increasing demand from secondary urban areas for higher-value crops. Both countries have made major progress in land titling, land-market development, and security of tenure, which has probably benefited women although formal statistics to confirm this are lacking. The RNFE is growing as well; faster in Tanzania because of higher incomes. Households in Tanzania are increasingly starting to specialise (another indicator of transformation), with members working full time either in farming or in non-farm activities. In both countries, extreme rural poverty is falling, although deprivation remains high.

There are clear signs of economic and social progress for women and girls in both countries. Especially in Tanzania, education and health indicators have improved: higher girls’ enrolment in schools; falling under-five mortality. Gender gaps in outcomes are closing. Ethiopia lags behind Tanzania on most social indicators by about 10–20 years. For example, serious nutrition issues persist in rural Ethiopia, affecting pregnant women and children, which will have long-term consequences. Women still work long hours doing household chores, especially in Ethiopia, where rural water is an issue and supply infrastructure lags.

Another contributor to the burden is high fertility. Ethiopia has made more progress than Tanzania on this front owing to a deliberate strategy to both reduce demand for large families and increase the supply of modern contraceptives. In Tanzania, half of women of childbearing age report no demand for contraception. This may be a women’s empowerment issue, but it may also be a traditional preference for large families combined with a lack of knowledge of options that has not been addressed in development strategy.

Overall, we do not see major changes in the norms that see a rural woman’s place as primarily a wife and mother, that assign rights to property first and foremost to men, and that tolerate worse treatment of women than men – including male domestic violence and abuse.

Progress for rural women, then, has been variable. Depending on which of the different dimensions of women’s empowerment are examined, it is possible to see promising

138

progress; or else see stasis punctuated by occasional, but limited, improvements in material conditions. It is hard to be more certain of the dynamics of changes to the lives of rural women and girls, partly owing to the limitations of the statistics available, and partly to the difficulty of judging future trajectories when often the key statistics available can only show changes over a relatively short period.

5.5.2 Prospects and policy leversWill the rural transformations continue, with the benefits widely shared? This is hard to predict. As the three historical reviews have shown, progress is rarely linear, and it is usually only with hindsight that the path can be clearly observed. Neither country appears to be in the situation of Peru 20 years ago – having a dualistic economic structure including a well-developed enterprise sector in urban and some coastal areas, rapidly falling fertility, and major progress in health and education for women and girls. In retrospect, Peru was ready for rural transformation once the country settled its conflict and decided to make the investments. The closest parallel country is probably Thailand in the 1970s. Could Ethiopia and Tanzania get Thailand-quality results in the next 30 years?

Tanzania is blessed with several advantages which should propel it into middle-income status: an Indian Ocean port providing cheap transport links to the Gulf States and Asia, and to four landlocked countries; offshore natural gas; and still underexploited agriculture and tourism potential. The government clearly needs to keep the focus on what brought the changes to date: an enabling rural investment climate, bolstered by agricultural research and social and economic infrastructure development. Family farms, smallholdings for the most part, can survive and grow if rural value chains continue to strengthen. This includes development of private input suppliers and buyers, supported by better rural financial services and continued improvement in transport and logistics infrastructure. Government policy has to avoid the over-regulation of the pre-2000 period, and allow private initiatives, such as mobile telephony and mobile financial services, to

flourish. It also has to improve public services, including in the energy sector and at the borders.

Although indicators of norm change within the household and community are limited, women appear to have high economic and social mobility, as they hold 40% of the wage jobs and an even higher share of RNFE employment. The best income-earning options for rural women include new areas such as horticulture and dairy, where they can earn an independent income, or the RNFE. More educated women can migrate to urban areas, where their skills will be in demand as private investment in the non-farm economy creates wage-earning opportunities. One troubling aspect is the lack of attention to lowering fertility and improving reproductive health. As seen in Thailand and Peru, this affects the lives of rural women and girls in many ways: through household labour supply as well as public and private savings and the ability of the public sector to improve the quantity and quality of services.

Ethiopia is much farther behind; for example, primary education enrolment is still not at the level that Thailand had in the mid-1970s, nor is GDP per person. But Ethiopia’s recent progress has been remarkable. The creation of export-processing zones (EPZs), which is attracting FDI to develop export-oriented manufacturing, could provide the non-farm output and employment boost Thailand had and Ethiopia needs, including agro-processing for export. As Africa’s second most populous country, Ethiopia could develop a robust non-farm enterprise sector in secondary cities to serve a growing domestic market. Ethiopia has several energy projects underway (hydro, geothermal, and solar) to ensure adequate energy supplies for the projected industrial growth. Ethiopia does not have a port, but it does have a recently opened rail link to Djibouti. Overall, Ethiopia has less medium- to high-potential arable land per person than Tanzania, however, suggesting that in some places households will have to leave rural areas to earn a better living. The question is how fast these opportunities will open up.

Progress for rural women and girls has been slow. But hopeful signs, such as the decline in early marriage and FGM/C, suggest that rigid norms can change. Much more investment in rural water

139

supply and education infrastructure is needed. If rural women obtain more education, and are not married early, they could benefit from the development of an export manufacturing sector. The jobs may be low-paying, hard work, but the income will belong to them, as will their free time. They will no doubt send remittances home. This experience will change their own attitudes as well as their families’, as it did in Thailand. For women who stay in rural areas, household livestock production and the RNFE offer the best opportunities.

Internal conflict has slowed Ethiopia’s progress before, and could do so again. If a political settlement can be reached in the long-simmering

ethnic conflicts, the government could devote more resources to development. Ethiopia needs foreign technology and capital to complement a small domestic enterprise sector. Services and prices in sectors which are not open to foreign investment, such as banking and telecoms, lag behind Ethiopia’s competitors. This, as well as the political environment, increase risks for private investors. As with Tanzania, the danger is in a public sector which likes to be in control, and has a low tolerance for the messiness of the informal economy or a private enterprise sector that could pose a political threat. Development requires a capable public sector, which enables rather than squashes private development initiatives.

140

6 Conclusions

This study examines the lives of rural women and girls within the broad context of economic development and structural transformation. This perspective recognises that much of what affects individuals is the outcome of collective endeavor on a large scale; be that through private firms and farms that produce goods and services; or in public services that enhance people’s capabilities; or in government that sets policies, laws and institutions that enable private and collective enterprise and protect the rights of individuals, households, and enterprises both private and collective.

These conclusions begin by recapping the main arguments.

6.1 Lessons from the country reviews

Egypt, Peru and Thailand since the early 1960s have each seen strong economic growth during which their economies have diversified away from agriculture. They have urbanised. In rural areas, agricultural output has grown well ahead of population growth; while the rural non-farm economy (RNFE) has flourished. Extreme poverty has almost disappeared; in Thailand only about 1% of the population lives on less than US$3 /day, while in Peru, owing to higher inequality, about 10% still do. Economic advances have created new and better earning opportunities for rural women, many of them off the farm, as well as raising their living standards.

Public services in rural areas have increased. Particularly important for rural women and girls has been increased provision of schooling, health facilities including maternal health care and primary health programmes, and family planning services. Most rural women have become much better educated than their grandmothers, enjoy better health and are much more likely to control their own fertility.

Hence the material conditions of life for most rural women in the three countries have been transformed. Less clear has been progress on gender equity: norms that privilege men to the detriment of women seem slower to improve than those in material conditions, but there has been progress in these as well.

Four main lessons can be derived from these experiences. One, broad-based development can generate many more opportunities to earn income to meet the rapid expansion in the labour force that arises in the early stages of a demographic transition. Moreover, these opportunities include activities with increased labour productivity and potentially with higher earnings.

Additional employment opportunities for rural people arise through various channels:

• in agriculture – largely on smallholdings in Egypt and Thailand, and in recent times in Peru, on larger-scale commercial holdings

• in the rural non-farm economy stimulated by increased agricultural incomes, which create demand for new goods and services from households, as well as investment in processing activities

• from public investment in education, health, and infrastructure

• in the transforming urban economy, which the rural population accesses either by commuting or migration.

Off-farm employment can be especially important for rural women. As farmers, many rural women operate at a disadvantage compared to men. Moreover, men may appropriate much of the produce and income they generate. Off-farm work for rural women may not only offer better returns than those available on the farm, but it can also generate cash that women can more readily control. Getting out of agriculture thus matters more for rural women than men.

141

Two, rural people can only take up better opportunities from economic growth if they are healthy, literate and numerate. Hence public investments in rural education, health, and clean water and sanitation make a real difference, and particularly for women and girls.

Three, of all rural services, family planning plays a special role for women. Access to advice and contraception put women in control of their fertility, almost always leading to fewer births and fewer children to care for. In addition, as population and labour force growth slow, labour shortages arise, investment to raise labour productivity is encouraged, and wages for unskilled work in rural areas start to rise, for both men and women.

Four, gender norms can either facilitate or block these processes. In these three countries, the freedom of rural women to travel independently stands out: rural women’s options widen when they can move to visit market centers, commute for jobs, or to migrate to the city. This shows up in higher earnings for working women as well as better outcomes for their children.

Ethiopia and Tanzania confirm the lessons from Egypt, Peru and Thailand on the importance of strong and broad-based economic development which includes the agricultural sector. The two African countries had followed the Peru of the 1970s and 1980s down the path of misguided economic policies, but as with Peru after 1990, eventually corrected earlier mistakes and began to achieve sustained economic expansion. Rural areas are transforming in both countries, with agricultural growth, more connections to markets, and growing rural non-farm economies. In both countries, extreme rural poverty is falling, although deprivation remains high. Both countries are still quite poor, however, and because the development of manufacturing has been limited, much of the diversification achieved to date has taken place within the household production sphere.

Both Ethiopia and Tanzania have increased public investment in rural services – health, education, water and sanitation. As a result, progress for rural women and girls is evident, in better health and more education, especially

in Tanzania. Gender gaps have been closing. That said, rural women still work long hours on household chores, especially in Ethiopia. Fertility remains high in both countries, declining faster in Ethiopia than Tanzania, with the latter country showing very small reductions indeed.

Few changes can be seen to norms that see a rural woman’s place as primarily a wife and mother, that assign rights to property first and foremost to men, and that tolerate worse treatment of women than men – including male domestic violence. Having said that, in both countries it is income as much as norms which restrict women’s freedom of movement. In Tanzania, when households have more money, qualitative and quantitative evidence suggests that they send their girls to the town or city for secondary education almost as often as their boys. The girls often stay there to work, sending money back to their families. Ethiopia may follow this trajectory as well, especially if the government’s industrialisation strategy works, and the export processing zones start offering wage-earning opportunities for more educated young women.

In summary, it is possible to see promising progress in Ethiopia and Tanzania; or to see stasis punctuated by occasional, but limited, improvements in material conditions. It is hard to be more certain of the dynamics of changes to the lives of rural women and girls, partly owing to the limitations of the statistics available, and partly to the difficulty of judging future trajectories when some of the key statistics available can only show changes over a relatively short period.

The longer-term studies of Egypt, Peru and Thailand suggest that it is only with the benefit of hindsight that remarkable progress can be seen. Some observers of rural Thailand in the 1980s, for example, doubted that sustained development that would favour ordinary rural people was underway (Ganjanapan and Hirsch, 2010). The sceptics focused on what had been important in the villages up until that time – access to land, rice production, for example – and failed to realise the significance of the growth of the rural non-farm economy, the proliferation of urban opportunities, girls going to school and falling rural fertility. The

142

question for the future trajectories in Ethiopia and Tanzania is whether the governments will continue to invest in broad-based development and better opportunities for rural households while at the same time laying the groundwork for new opportunities in urban areas.

Looking across the five countries, see Table 5 for a summary of changes, several points are common to all or most of the country experiences.

One is that progress for rural women tends to be uneven: more has been achieved in creating economic opportunities and in providing the services that allow human development, than in changing women’s roles in the household or in status and agency. Gender norms apply more tightly to the latter two dimensions, than to the former two.

This prompts the question of whether gender norms change more in response to legal actions to create and reinforce women’s rights and protect them from abuse; or to economic growth that creates more economic opportunities combined with increased provision of health, and education and infrastructure services. In the case of the five countries, improvements in women’s status seems to be more a result of their enhanced ability to earn income, to read, write, and control their fertility, than from laws and regulations. Moreover, gender norms that limit women seem to have shifted more to accommodate economic opportunity, schooling and health, than in women’s roles in households and villages.

This is not to argue that direct legal measures are not needed. Moreover, any optimism that enhanced incomes, education and health for rural women would transform their status has to be moderated in the light of the limited progress towards gender equality seen across the countries. Clearly progress can be made by either approach, but it does raise questions about priorities and sequences.

6.2 Looking forward. How can agricultural development help rural women and girls?The original question posed for this study concerned what an inclusive agricultural transformation that empowers women would be. Our study, however, arrives at a stark conclusion: the lives of most rural women and girls improve

most as they move out of agriculture. Hence, the conditions that lead to the shift of the labour force out of farming have taken precedence over discussions of forms of agricultural development.

This should not, however, be interpreted as meaning that agricultural development is not important. How could it not be? The majority of rural households in developing countries are still, in the second decade of the 21st century, engaged to some degree in farming.

Moreover, the growth of agricultural production and productivity, well ahead of population growth, has been central to the development of all five countries reviewed. Growing agricultural incomes create demand for many of the goods and services supplied in the rural non-farm economy. When agricultural growth faltered in Ethiopia, Peru and Tanzania through much of the 1970s and 1980s, the result was a severe drag on overall economic development, and deep poverty among the rural population. Timmer’s adage (2009) that the fastest way for a country to reduce its dependence on agriculture is through agricultural growth is confirmed in the experiences reviewed.

Hence any low-income, agrarian country seeking to improve the lives of rural women and girls has to promote agricultural development. Broad-based agricultural development that includes smallholders is both feasible and desirable in that it tends to raise rural income faster than growth based on large-scale farming. The experience of Peru, however, reminds that judgments about scale of enterprise need to be pragmatic: the export agriculture of large farms on the coast has generated much-needed jobs that have finally started to tighten rural labour markets – to the advantage of female migrant workers.

The last ten or so years have seen an awakening of interest in women in agriculture. This has not been in vain. Far from it. Most women farmers are disadvantaged and much can be done to remedy this – in strengthened women’s rights to land, better and fairer access to technical advice and marketing, inputs, and services. The agenda established (see, for example, FAO, IFAD and ILO, 2010; World Bank and ONE, 2014) deserves its rising prominence; but it needs to be accompanied by wider development measures.

143

Economic participants Mothers, wives and carers

Women’s human development

Citizens and agents

Thailand Status: Rural women were able to work in thriving RNFE, migrate to urban area for better opportunities

Status: Low fertility rates reduced number of children, and widespread access to running water and electricity all reduced the burden of care

Status: High rates of girls schooling, improved health

Status: Norms allowed women to move, to undertake business — but women remain under-represented in political leadership and business

Drivers: Economic growth and transformation, education, norms permit migration

Drivers: Highly effective family planning program since early 1970s, public investments in water and power

Drivers: Public investment in health and education, norms in favour of education

Drivers: Rural social norms gradually blending in urban ideas about modernity

Peru Status: Rural women gained opportunities on large commercial farms, in the RNFE, and by migrating to cities

Status: Falling fertility and some improvements in access to water, electricity reduced the care burden; public child care services for some rural women

Status: Higher rates of girls’ secondary schooling, moderate improvements in rural health

Status: in hierarchical society with marked class and ethnic divisions, most rural women have low status from intersection of gender with class and ethnicity. High rates of violence against women

Improvements limited to date.

Drivers: Economic growth, especially labor-intensive vegetable fields

Drivers: Public investments in rural services including child care

Drivers: Public investments in rural schools, health centres; changing norms on rural girls’ schooling as jobs apparent for women

Drivers: laws to protect women, services for women, quotas for seats in local government have been brought in — but change encounters deep traditions favouring men

Egypt Status: Rural women only allowed to work in home and nearby fields. Opportunities for educated women in public sector declined; sluggish private sector labor demand favoured men.

Status: Modest reduction in fertility and housing improvements have eased care demands

Status: Rising rates of girls’ secondary schooling, but much less so for those from low-income households. Health improvements

Status: legally women have many rights, but in custom and practice they have limited property rights and limited freedom to leave the household unaccompanied

Drivers: Gender norms impeded women moving alone, economy not generating enough employment

Drivers: Public investment in rural water and electricity

Drivers: public investments in schools and health services

Drivers: gender norms privilege men, with only small changes in recent times.

Ethiopia Status: Some increased opportunities, but limited development of the rural non-farm economy — and of the urban economy

Status: Some small reductions in fertility but accelerating downward trend, better access to water; no major change in care burden seen yet

Status: Considerable improvements to rural health, many more girls going to primary school, but rural secondary school enrolment by girls (and boys) is very low

Status: Legal provisions for gender equality exist, but implementation stymied by rural norms that subordinate women to men, and tolerate violence against women.

Land certification has strengthened women’s rights to property

Drivers: Strong agricultural growth, albeit from a low base

Drivers: Public investments in rural services; public commitment to smaller families

Drivers: Public spending on health and education, although too few rural secondary schools

Drivers: Formal laws and land certification, offset by difficulties of implementing in rural areas when formal rules clash with patriarchal gender norms

Table 5 Summarising changes in four dimensions of rural women’s lives in the five countries

144

Economic participants Mothers, wives and carers

Women’s human development

Citizens and agents

Tanzania Status: More opportunities, mainly in the rural non-farm economy

Status: high fertility; marginal improvements in water supply and electricity mean few changes yet in broad segment of the country

Status: Improved access of girls to schools; better health

Status: Some improvement in women’s status, but formal legal provisions do not readily translate to improved gender norms

Drivers: Strong economic growth, including agricultural development and the rural non-farm economy

Drivers: No commitment to reducing high rates of fertility, encouraging reproductive health mean large families

Drivers: Public spending on health and education

Drivers: Increase earnings and education by some women have enhanced status

145

References

Abdel Aal, M. (2002) Women’s Participation in Agricultural Activities, Background Paper, http://www.un.org/womenwatch/daw/csw/csw56/egm/Martini-EP-9-EGM-RW-Sep-2011.pdf

Adireksombat, Kampon, Fang Zheng and Chris Sakellariou (2010) The Evolution of Gender Wage Differentials and Discrimination in Thailand: 1991-2007 — An Application of Unconditional Quantile Regression, Working Paper 2010/05, Economic Growth Centre, Nanyang Technological University, Singapore

Admassie, A., Nuru, S., May, J., and Megquier, S. (2015) The demographic dividend: An opportunity for Ethiopia’s transformation. Addis Abeba: Ethiopian Economic Association.

Agricultural Transformation Agency (ATA). (2016, 12 19) Gender Equality. Addis Abeba: Agricultural Transformation Agency.

Akkari, A. (2004) Education in the Middle East and North Africa: The Current Situation and Future Challenges. International Education Journal, 5(2), 144–153.

Al Monitor (2016) Inheritance plan could be game change for Egyptian women, July 11 2016. Al Monitor. Available online at: http://www.al-monitor.com/pulse/originals/2016/07/egypt-inheritance-rights-women-disputes-courts.html

Alemu, G. (2014) Financial inclusion, regulation and inclusive growth in Ethiopia. London: Overseas Development Institute.

Alemu, G. T. (2015) Women’s land use right policy and household food security in Ethiopia: Review. International Journal of African and Asian Studies, 12, 56–65.

Alesina, A., Giuliano, P., and Nunn, N. (2013) On the origins of gender roles: Women and the plough. Quarterly Journal of Economics, 469–530.

Ali, Daniel, Deininger, K. and Harris, A. (2016) Large Farm Establishment, Smallholder Productivity, Labor Market Participation, and Resilience. Evidence from Ethiopia. Policy Research Working Paper 7576. Washington, DC: World Bank.

Alkire, S., and Santos, M. (2010) Ethiopia Country Briefing. Oxford: University of Oxford.Alkire, S., Meinzen-Dick, R., Peterman, A., Quisumbing, A., Seymour, G., and Vaz, A. (2013) The

Women’s Empowerment in Agriculture Index. World Development 52: 71–91.Alkire, S., Roche, J., and Vaz, A. (2014) Multidimensional Poverty Dynamics: Methodology and

Results for 34 countries. Oxford: Oxford Policy and Human Development Initiative (OPHI).Altamirano, T. (2003) From Country to City. Internal Migration, ReVista. Harvard Review of Latin

America [http://revista.drclas.harvard.edu/book/country-city-english-version]Alvergne, A., Gibson, M., Gurmu, E., and Mace, R. (2011) Social transmission and the spread of

modern contraception in rural Ethiopia. PLoS One, online.Ambel, A., Mehta, P., and Yigezu, B. (2015) Multidimensional Poverty in Ethiopia. Changes in

overlapping deprivations. Washington, DC: World Bank.Ames, P. (2013) Constructing new identities? The role of gender and education in rural girls’ life

aspirations in Peru. Gender and Education, 25(3), 267–283. https://doi.org/10.1080/09540253.2012.740448

Amin, S., and Lloyd, C. B. (2002) Women’s Lives and Rapid Fertility Decline: Some Lessons from Bangladesh and Egypt. Population Research and Policy Review, 21(4), 275–317. https://doi.org/10.1023/A:1020030616075

Amnesty International (2012) Stifling human rights work. The impacts of civil society legislation in Ethiopia. London: Amnesty International.

Andersson, Jens A. (1996) Potato cultivation in the Uporoto mountains, Tanzania. An analysis of the social nature of agro-technological change. African Affairs, 95, 85–106

146

Assaad, R. (2002) Informalization and De-Feminization: Explaining the Unusual Pattern in Egypt. In Conference on Rethinking Labor Market Informalization: Precarious Jobs, Poverty, and Social Protection, Cornell University, Ithaca, NY, October (pp. 18–19). Retrieved from http://wwwhhh.oit.umn.edu/people/rassaad/pdf/informalization.pdf

Assaad, R. (2004) Why Did Economic Liberalization Lead to Feminization of the Labor Force in Morocco and De-Feminization in Egypt. In Center of Arab Women Training and Research, Mediterranean Development Forum. Retrieved from https://www.researchgate.net/profile/Ragui_Assaad/publication/228376929_Why_Did_Economic_Liberalization_Lead_to_Feminization_of_the_Labor_Force_in_Morocco_and_De-feminization_in_Egypt/links/00b7d5294cdb605ebf000000.pdf

Assaad, R., and Arntz, M. (2005) Constrained Geographical Mobility and Gendered Labor Market Outcomes Under Structural Adjustment: Evidence from Egypt. World Development, 33(3), 431–454. https://doi.org/10.1016/j.worlddev.2004.08.007

Assaad, R., El-Hamidi, F., Ahmed, A. U., and others. (2000) The Determinants of Employment Status in Egypt (No. FCND Discussion Paper No. 88). Washington D.C.: International Food Policy Research Institute. Retrieved from http://ageconsearch.umn.edu/bitstream/16440/1/fc000088.pdf

Assad, R., and Barsoum, G. (2007) Youth Exclusion in Egypt: In Search of “Second Chances.” Middle East Youth Initiative Working Paper, (2). Retrieved from http://papers.ssrn.com/sol3/papers.cfm?abstract_id=1087429

ATA (2015) GTP2 and the Agricultural Transformation Agenda. Addis Abeba: Agricultural Transformation Agency.

Atnafu, A., Oucho, L., and Zeitlyn, B. (2014) Poverty, youth and rural-urban migration in Ethiopia. Brighton: University of Sussex.

Bachewe, F., Berhane, G., Minten, B., and Taffesse, A. (2015) Agricultural Growth in Ethiopia (2004-2014): Evidence and drivers. Washington, DC and Addis Abeba: IFPRI and EDRI.

Bachewe, F., Berhane, G., Minten, B., and Taffesse, A. (2016) Non-farm income and labour markets in rural Ethiopia. Washington, DC and Addis Abeba: IFPRI and EDRI.

Bantje, H. (1986) Household differentiation and productivity. A study of smallholder agriculture in Mbozi District. Research Paper No 14. Dar Es Salaam: Institute of Resource Assessment, University of Dar Es Salaam.

Barrett, C. B., Reardon, T. and Webb, P. (2001) Nonfarm Income Diversification and Household Livelihood Strategies in Rural Africa: Concepts, Dynamics, and Policy Implications. Food Policy 26 (4): 315–331.

Barrios, E. (2015) Surviving Juntas (together): Lessons of Resilience of Indigenous Quechua Women in the Aftermath of Conflict in Peru.

Berhane, G., Hoddinott, J., Kumar , N., Seyoum , A., Diressie, M., Yohannes, Y., and Sima, F. (2013) Evaluation of Ethiopia’s Food Security Program: Documenting Progress in the Implementation of the PSNP and HABP. Washington, DC: IFPRI.

Bertrand, M., and Pan, J. (2013) The trouble with boys: Social influences and the gender gap in disruptive behavior. American Economic Journal: Applied Economics, 5(1), 32-64.

Bevan, P., Dom, C., and Pankhurst, A. (2010) Long term perspectives on development impacts in rural Ethiopia. WIDE Stage One Final Report. Oxford: Mokoro.

Bevan, P., Dom, C., and Pankhurst, A. (2013) Long-term perspectives in development impacts in rural Ethiopia: WIDE 3 Stage 2 Final Report. Oxford: Mokoro.

Bevan, P., Dom, P. and Pankhurst, A. (2014) Long term perspectives on development impacts in rural Ethiopia: WIDE3 stage 3. Six communities with agricultural potential: short summary. Oxford: Mokoro.

Beyene, H. (2015) Final report national assessment: Ethiopia gender equality and the knowledge society. Los Angeles, CA: Swedish International Development Cooperation Agency, Women in global science and technology.

147

Bezabih, M., and Holden, S. (2010) The role of land certification in reducing gender gaps in productivity in rural Ethiopia. Gothenburg and Washington, DC: University of Gothenburg and Resources for the Future.

Bezabih, M., Holden, S. and Mannberg, A. (2016) The role of land certification in reducing gaps in productivity between male-and female-owned farms in rural Ethiopia. Journal of Development Studies, 52(3), 360–376.

Bezu, S., and Holden, S. (2014) Are rural youth in Ethiopia abandoning agriculture? World Development, 64, 259–272.

Bezu, S., Barrett, C., and Holden, S. (2012) Does the nonfarm economy offer pathways for upward mobility? Evidence from a panel data study in Ethiopia. World Development, 1634–1646.

Binzel, C., and Assaad, R. (2011) Egyptian men working abroad: Labour supply responses by the women left behind. Labour Markets in Developing Countries, 18, S98–S114. https://doi.org/10.1016/j.labeco.2011.03.002

Bishop, C. (2017) Striving for gender transformative impacts in rural livelihoods. Background paper, UN Women, IFAD, FAO, WFP Expert Group Meeting, ‘Challenges and opportunities in achieving gender equality and the empowerment of rural women and girls, Rome, Italy, 20-22 September 2017.

Boesen, J. and Ravnborg, H.M. (1992) Peasant production in Iringa District, Tanzania. CDR Project Paper 93.1, Centre for Development Research, Kbh, DK.

Boesten, J. (2012) The State and Violence Against Women in Peru: Intersecting Inequalities and Patriarchal Rule, Social Politics, 19 (3), 361–382. http://sp.oxfordjournals.org/cgi/doi/10.1093/sp/jxs011

Boyden, J., Pankhurst , A., and Tafere, Y. (2013) Harmful traditional practices and child protection: Contested understandings and practices of female child marriage and circumcision in Ethiopia. Oxford: Young Lives, University of Oxford.

Brady, M. and Population Council. (2007) Providing New Opportunities to Adolescent Girls in Socially Conservative Settings: The Ishraq Program in Rural Upper Egypt. New York: Population Council.

Brockington, D. (2018) Longterm Livelihood Change in Tanzania. http://livelihoodchangeta.wixsite.com/tanzania

Brody, C., de Hoop, T., Vojtkova, M., Warnock, R., Dunbar, M., Murthy, P., and Dworkin, S. (2015) Economic self-help group programs for improving women’s empowerment: A systematic review. Oslo: The Campbell Collaboration.

Bryant, J. and Rossarin, G. (2005) Rural population ageing and farm structure in Thailand. Population and Development Service Sustainable Development Department. Food and Agriculture Organisation of the United Nations.

Butterworth, J., Welle, K., Hailu, T., Bostoen, K., and Schäfer, F. (2013) Monitoring access to rural water supplies in Ethiopia. A background paper to the National WASH Inventory seminar, Addis Abeba. The Hague: IRC.

Byker, T., and Gutierrez, I. A. (2012) Fertility and Family Well-being Effects of an Aggressive Family Planning Policy in Peru in the 1990s. Population Studies Centre Research Reports 12-765, Ann Arbor, MI: University of Michigan.

Cain, M., Khanam, S. R., and Nahar, S. (1979) Class, patriarchy, and women’s work in Bangladesh. Population and Development review, 405-438.

Calow, R., Mason, N., and Ludi, E. (2016) WASH Literature Review in support of WASH impact review commissioned by ICAI. Unpublished.

CFSVA (2014) Ethiopia Comprehensive Food Security and Vulnerability Analysis. Addis Abeba: Ethiopia Central Statistical Agency and World Food Programme.

Chamberlin, J., and Schmidt, E. (2011) Ethiopian Agriculture: A dynamic geographic perspective. Washington, DC and Addis Abeba: IFPRI and EDRI.

Chamratrithirong A; Archavanitkul K; Richter K; Guest P; Thongthai V; Boonchalaksi W; Piriyathamwong N; Vong-Ek P. (1995) National migration survey of Thailand. Nakhon Pathom, Thailand: Mahidol University, Institute for Population and Social Research Publication No. 188

148

Chen, M., Vanek, J., Lund, F., and Heintz, J. (Eds.). (2005) Progress of the World’s Women 2005: Women, Work, and Poverty. New York, N.Y: United Nations Development Fund for Women.

Clausen, A. (2004) Rural-urban divide or continuum?: on the consequences of female labour migration to Bangkok for rural-urban interactions in an age of globalisation. Geographica Helvetica, 59(1), 44-53

Croppenstedt, A. (2006) Household Income Structure and Determinants in Rural Egypt. Retrieved from http://www.fao.org/3/2cc4d4c3-1bbd-5978-b1c0-67b5b6e33ffc/af840e00.pdf

CSA and World Bank (2013) Ethiopia Rural Socio-economic Survey (ERSS). Addis Abeba and Washington, DC: CSA and World Bank.

CSA and World Bank (2015) Ethiopia Socio-economic Survey (ESS), 2013-14. Addis Abeba and Washington, DC: CSA and LSMS, World Bank.

CSA and ICF (2012) Ethiopia Demographic and Health Survey 2011. Addis Abeba and Calverton, MD: CSA and ICF.

CSA and ICF (2016) Ethiopia – Demographic and Health Survey 2016. Key Indicators Report. Addis Abeba and Rockville: Central Statistical Agency and ICF.

CSA, and ORC Macro (2001) Ethiopia Demographic Health Survey 2000. Addis Abeba and Calverton, MD: CSA and ORC Macro.

CSA, and ORC Macro. (2006) Ethiopia Demographic and Health Survey 2005. Addis Abeba and Calverton, MD: CSA and ORC Macro.

CSA, NBE, and World Bank (2017) LSMS - Integrated Surveys on Agriculture Ethiopia Socioeconomic Survey. Addis Ababa: CSA, NBI, World Bank.

Cupito, E., and Langsten, R. (2011) Inclusiveness in higher education in Egypt. Higher Education 62:183–197. Available online at: https://www.researchgate.net/profile/Ray_Langsten/publication/225657292_Inclusiveness_in_higher_education_in_Egypt/links/55364d350cf268fd0016bc76.pdf

De Magalhaes, Leonardo and Raul Santaeulalia-Llopis (2015) The Consumption, Income and Wealth of the Poorest. World Bank Policy Working Paper 7337, June, 2015.

Deininger, K., Ayalew Ali, D., and Alemu, T. (2008) Impacts of Land certification on Tenure Security, Investment and Land Markets. Evidence from Ethiopia. Washington, DC: World Bank.

Deininger, K., Ayalew Ali, D., Holden, S. and Zevenbergen, J. (2007) Rural land certification in Ethiopia: Process, initial impact, and implications for other African countries. World Bank Policy Research Working Paper 4218. Washington, DC: World Bank.

Delgado, C. L., and Minot, N.W. (2000) Agriculture in Tanzania since 1986: follower or leader of growth? Washington, DC: World Bank.

Denney, L., and Kassaye, D. (2013) Securing communities for development: Community policing in Ethiopia’s Amhara National Regional State. London: Overseas Development Institute.

Derbew, D. (2013) Ethiopia’s renewable energy potential and development opportunities. Addis Abeba: Ministry of Water and Energy.

Dercon, S. and Zeitlin, A. (2009) Rethinking Agriculture and Growth in Ethiopia: A Conceptual Discussion. Paper prepared as part of a study on Agriculture and Growth in Ethiopia. Oxford: Centre for Study of African Economies.

Dercon, S. (1994) Peasant supply response and macroeconomics policies: cotton in Tanzania. Journal of African Economies, 2 (2), 157–93.

Dixon, C. (1999) The Thai economy. Uneven development and internationalism, Routledge, LondonDolado, J.J., Felgueroso, F., and Jimeno-Serrano, J.F. (2002) Recent Trends in Occupational

Segregation by Gender: A Look across the Atlantic. http://papers.ssrn.com/sol3/papers.cfm?abstract_id=321092

Donald, A. Koolwal, G., Annan, J., Falb, K., and Goldstein, M. (2016) Measuring Women’s Agency. http://pubdocs.worldbank.org/pubdocs/publicdoc/2016/6/487641466185525738/2016-05-31-Measuring-Womens-Agency.pdf

149

Dorosh, P. and Hashim, A. (2009) Foreign Exchange Rationing and Wheat Market Distortions, and Food Security in Ethiopia. Paper presented at Ethiopian Economics Association, Seventh International Conference on the Ethiopian Economy, 26 June, Addis Ababa.

Doss, C. (2013) Intrahousehold bargaining and resource allocation in developing countries. The World Bank Research Observer, 28 (1), 52–78.

Drake, J.K., Vail, J.G. Stewart, J.R. (2014) Best practices for fostering family planning total markets in Indonesia, Mexico, Romania, Thailand, and Turkey: a retrospective analysis. Cases in Public Health Communication and Marketing 8(suppl 1): S19-S41.

Duarte, M., and Restuccia,D. (2010) The role of the structural transformation in aggregate productivity. The Quarterly Journal of Economics, 125(1), 129–73. doi:10.1162/qjec.2010.125.1.129.

Ekasingh, B., Sungkapitux, C., Kitchaicharoen, J., and Suebpongsang, P. (2007) Competitive Commercial Agriculture in the Northeast of Thailand. The World Bank, Washington, DC.

El-laithy, H., Lokshin, M., and Banerji, A. (2003) Poverty and Economic Growth in Egypt, 1995-2000 (SSRN Scholarly Paper No. ID 636430). Rochester, NY: Social Science Research Network. Retrieved from https://papers.ssrn.com/abstract=636430

Ellis, F. (1983) Agricultural marketing and peasant–state transfers in Tanzania. The Journal of Peasant Studies, 10 (4), 214–242.

EPHI (2013) Countdown to a healthier Ethiopia: Building on successes to accelerate newborn survival. Addis Abeba: Ethiopian Public Health Institute.

Ersado, L. (2012) Education in Egypt: inequality of opportunity across three generations, World Bank Blog, 12/18/2012, https://goo.gl/mCn8rL

Espinoza Benza, A. (2016) La situación de la mujer rural en el Perú. MimeoFan, S., Jitsuchon, S. and Methakunnavut, N. (2004) The importance of public investment for

reducing rural poverty in middle-income countries: the case of Thailand, DSGD Discussion Paper 7, Washington DC, International Food Policy Research Institute.

FAO (2011) The State of Food and Agriculture 2010/11. Women in agriculture. Closing the gender gap for development. Rome: Food and Agriculture Organisation of the United Nations.

FAO LINKS (2003) Livestock and Gender: The Tanzanian Experience in Different Livestock Production Systems. Links project case study. http://www.fao.org/fileadmin/templates/esw/esw_new/documents/Links/Publications_Tanzania/17_CS_3_Livestock_and_Gender.pdf

FAO, IFAD and ILO (2010) Gender dimensions of agricultural and rural employment: Differentiated pathways out of poverty. Status, trends and gaps. Rome: Food and Agriculture Organisation of the United Nations.

FDRE (1995) Constitution of the Federal Democratic Republic of Ethiopia. Addis Abeba: FDRE.Ferede, T. and Kebede, S. (2015) Economic growth and employment patterns, dominant sector, and

firm profiles in Ethiopia: Opportunities, challenges and Prospects. R4D Working Paper 2015/2. Berne: World Trade Institute, University of Berne.

Ferede, T., and Kebede, S. (2015) Economic growth and employment patterns, dominant sector, and firm profiles in Ethiopia: Opportunities, challenges and prospects. Berne: World Trade Institute, University of Berne.

Filmer, D. and Fox, L. (2014) Youth Employment in Sub-Saharan Africa. Washington, DC: World Bank. http://elibrary.worldbank.org/doi/book/10.1596/978-1-4648-0107-5.

Fort, R. and Paredes, H. (2015) Inversión pública y descentralización: sus efectos sobre la pobreza rural en la última década. Lima: GRADE.

Fox, L. and Thomas, A. (2016) Africa’s got work to do: A diagnostic of youth employment challenges in Sub-Saharan Africa. Journal of African Economies, 25 (suppl 1), i16–36. doi:10.1093/jae/ejv026.

Fox, L. (2015) Will Women in Low-Income Countries Get Lost in Transformation? London: Overseas Development Institute.

150

Frazer, E. (1992) Cooperative partnerships, innovative marketing, and a sense of fun have helped Thailand defuse the population bomb, In Context 31.

Funahashi, K. (1996) Farming by the older generation: The exodus of young labor in Yasothon province, Thailand. Southeast Asian Studies 33(4), 107-21.

Funk, C., Rowland, J., Eilerts, G., Kebede , E., Biru, N., White, L., and Galu, G. (2012) A Climate Trend Analysis of Ethiopia. Reston, Washington, DC and Addis Abeba: USGS, USAID and FewsNet.

Ganjanapan, A. and Hirsch, P. (2010) Transforming Agrarian Transformation in a Globalizing Thailand. In Hirsch, P. and Tapp, N. (eds). Tracks and Traces. Thailand and the Work of Andrew Turton. Amsterdam: Amsterdam University Press.

Ganjanapan, A. (1994) The Northern Thai land tenure system: local customs versus national laws. Law and Society Review, 609-622.

Garip, F., and Curran, S. (2010) Increasing migration, diverging communities: changing character of migrant streams in rural Thailand. Population research and policy review 29(5), 659-685.

Gebeyehu, Z. (2014) Rural–urban migration and land and rural development policies in Ethiopia. 2014 World Bank Conference on Land and Poverty. Washington, DC: World Bank.

Gella, A., and Tadele, G. (2014) Gender and farming in Ethiopia: an exploration of discourses and implications for policy and research. Brighton: Future Agricultures.

Ghanem, H. (2015) Egypt’s Difficult Transition: Options for the International Community. In The Arab Spring Five Years Later: Case Studies (pp. 1–47). Retrieved from https://goo.gl/xXvnb1

Girma, H., and Giovarelli, R. (2013) Gender implications of joint land titles in Ethiopia. Focus on Land in Africa. http://www.focusonland.com/download/5719f54f21d05/

Githinji, M. and Mersha, G. (2007) Untying the Gordian knot. The question of land reform in Ethiopia. In Haroon Akram-Lodhi, A., Borrass Jr, S. and Kay, C. (Eds). Land, Poverty and Livelihoods in an Era of Globalization. Perspectives from developing and transition countries. London and New York: Routledge.

GNESD. (2018) Accelerated Rural Electrification Programme of Thailand. http://energy-access.gnesd.org/cases/16-accelerated-rural-electrification-programmeme-of-thailand.html

Government of Egypt (2014) Constitution of the Arab Republic of Egypt. (Unofficial translation). State Information Service, Cairo. Available online at: http://www.sis.gov.eg/Newvr/Dustor-en001.pdf

Grove, A. (1993) Water use by the Chagga on Kilimanjaro. African Affairs, 92, 431–448.Guinand, Y. (forthcoming). Policy Issues and Outlook. In Hurni, H. (ed.). Soil Erosion and

Conservation in Ethiopia. Berne: Centre for Development and Environment.Haggblade, S., Hazell, P, and Reardon, T. (2010) The rural non-farm economy: Prospects for growth

and poverty reduction. World Development, 38 (10), 1429–1441.Haggblade, S., Hazell, P.B.R, and Reardon, T. (2007) Transforming the Rural Nonfarm Economy:

Opportunities and Threats in the Developing World. Baltimore, MD: Johns Hopkins University Press.

Hallward-Driemeier, M., and Gajigo, O. (2015) Strengthening economic rights and women’s occupational choice: The impact of reforming Ethiopia’s family law. World Development, 260–273.

Hassan, M., and Sassanpour, C. (2008) Labor Market Pressures in Egypt: Why Is the Unemployment Rate Stubbornly High? Journal of Development and Economic Policies, 10(2), 77–115.

Head, S., Zweimueller, S., Marchena, C., and Hoel, E. (2014) Women’s lives and challegnes: Equality and empowerment since 2000. Rockville, MD: ICF International.

Herrendorf, B., Rogerson, R., and Valentinyi, A. (2013) Growth and Structural Transformation. w18996. Cambridge, MA: National Bureau of Economic Research. http://www.nber.org/papers/w18996.pdf

Herrmann, R., Nasarudin, S., Manfred, W. (1989) How non-agricultural import protection taxes agricultural exports: a true protection: analysis for Peru and Malaysia, Kiel Working Paper, No. 394, Institut für Weltwirtschaft (IfW), Kiel.

151

Holden, S., and Tefera, T. (2008) Land Certification in Ethiopia – Early Impacts on Women. United Nations Human Settlements Programme. Nairobi: UN Habitat.

Holmes, R., and Jones, N. (2011) Gender inequality, risk and vulnerability in the rural economy. Refocusing the public works agenda to take account of economic and social risks. Rome: Food and Agriculture Organisation of the United Nations.

Huguet, J., Chamratrithirong, A., and Natali, C. (2012) Thailand at a crossroads: Challenges and opportunities in leveraging migration for development. Issue in brief-IOM Regional Office for Asia and the Pacific and the Migration Policy Institute, 6, 1-12.

Hydén, G. (1980) Beyond Ujamaa in Tanzania: underdevelopment and an uncaptured peasantry. Oakland, CA: University of California Press.

IFAD (2016) Rural Development Report 2016: Fostering inclusive rural transformation. Rome: International Fund for Agricultural Development.

IMF (2016) The Federal Democratic Republic of Ethiopia. 2016 Article IV Consultation - Press Release. Washington, DC: International Monetary Fund.

Inchauste, G., Olivieri, S., Saavedra-Chanduvi, J., and Winkler, H. (2012) What is behind the decline in poverty since 2000? evidence from Bangladesh, Peru and Thailand. World Bank Policy Research Working Paper 6199 (2012), Washington DC: World Bank

International Fund for Agricultural Development (2006) Arab Republic of Egypt. Country Strategic Opportunities Paper, Executive Board – Eighty-seventh Session, Rome, 19-20 April 2006.

Isvilanonda, S., Ahmad, A. and Hossain, M. (2001) Recent Changes in Thailand’s Rural Economy: Evidence from Six Villages, Economic and Political Weekly, 35 (52.53), 4644–4649.

Jansen, K. and Khannabha, C. (2009) The Fiscal Space of Thailand: A Historical Analysis. Fiscal Space, Policy Options for Financing Human Development, 325-89.

Jansen, K. (2001) Thailand: The making of a miracle? Development and Change, 32(2), 343-370.Jasper, C., Thanh-Tam, L., Bartram, J. (2012) Water and Sanitation in Schools: A Systematic Review of

the Health and Educational Outcomes. Int. J. Environ. Res. Public Health 9(8), 2772–2787.Jitramontree, P. (2007) Thailand‘s government support for Family Planning. Columbia University

Journal of Bioethics and Supplement on BIOCEP, 79.Johnson, N., Balagamwala, M., Pinkstaff, C., Theis, S., Meinzen-Dick, R., and Quisumbing, A. (2017)

How do agricultural development projects aim to empower women? Insights from analysis of project strategies. Washington, DC: IFPRI.

Johnston, B. F., and Mellor, J. W. (1961) The role of agriculture in economic development. The American Economic Review, 51(4), 566-593.

Johnston, B.F. and Kilby, P. (1975) Agriculture and Structural Transformation. New York: Oxford University Press.

Jones, N., Emirie, G., Tefera, B., and Presler-Marshall, E. (2016) Surprising trends in child marriage in Ethiopia. London: ODI and Unicef.

Joseph, S., and Slyomovics, S. (Eds.). (2011) Women and power in the Middle East. University of Pennsylvania Press.

Karimli, L., Samman, E., Rost, L., and Kidder, T. (2016) Factors and norms influencing unpaid care work. Household survey evidence from five rural communities in Colombia, Ethiopia, the Philippines, Uganda and Zimbabwe. Oxford: Oxfam GB.

Kasa, L., Abate, G., Warner, J., and Kieran, C. (2015) Patterns of agricultural production among male and female holders: Evidence from Agricultural Sample Surveys in Ethiopia. Washington, DC: IFPRI.

Kay, C. (1982) Achievements and contradictions of the Peruvian agrarian reform. Journal of Development Studies, 18 (2), 141–170.

Khodair, A. and Hassib, B (2015) Women’s Political Participation in Egypt: The Role of the National Council for Women. Available online at: https://www.psa.ac.uk/sites/default/files/conference/papers/2015/Women’s%20political%20participation%20in%20Egypt.pdf

152

Kirigia, E., Betsema, G., van Westen, G., and Zoomers, A. (2016) Flowers for food? Scoping study on Dutch flower farms, land governance and local food security in East Africa. Utrecht: Landac/IDS, Utrecht University.

Korten, D.C. (1980) Community organisation and rural development: a learning process approach, Public Administration Review, Sept/Oct 1980.

Krongkaew, M. (1985) Agricultural development, rural poverty, and income distribution in Thailand. The Developing Economies, 23(4), 325-346.

Krongkaew, M., Chamnivickorn, S. and Nitithanprapas, I. (2006) Economic growth, employment, and poverty reduction linkages: the case of Thailand. Issues in Employment and Poverty Discussion Paper #20, Geneva: ILO (January). Download from: http://ilo-mirror.library.cornell.edu/public/english/employment/recon/poverty/download/disc20.pdf

Krueger, A., Schiff, M., and Valdés, A. (1991) The Political Economy of Agricultural Pricing Policy. Baltimore, MD and London: Johns Hopkins University Press.

Kumar, N., and Quisumbing, A. (2014) Policy reform toward gender equality in Ethiopia: Little by little the egg begins to walk. World Development, 67, 406–423.

Kweka, J., and Fox, L. (2011).The household enterprise sector in Tanzania: why it matters and who cares. Policy Research Working Paper 5882, Washington DC: World Bank

Laosutsaen, P. (2017).Thailand’s Migration Survey. Presentation of initial results. Bangkok: National Statistical Office.

Larson, J. M., Smith, S.M., Abler, D.G., and Trivelli, C. (2000) Land Titling In Peru: Is It Fulfilling Its Promise?. In 2000 Annual meeting, July 30–August 2, Tampa, FL, no. 21809. American Agricultural Economics Association (New Name 2008: Agricultural and Applied Economics Association).

Lavers, T. (2017) Land registration and gender equality in Ethiopia: How state–society relations influence the enforcement of institutional change. Journal of Agrarian Change, 188–207.

Lavigne, M. (2013) Social protection systems in Latin America and the Caribbean. Peru, Santiago de Chile: CEPAL.

Lemma, M., and Tesema, E. (2016) New approaches and methods for addressing gender gaps in extension services: experiences and lessons from the LIVES project in Ethiopia. Journal of Agricultural Economics, Extension and Rural Development, 429–435.

Lenhardt, A., Rogerson, A., Guadagno, F., Berliner, T., Gebreeyesus, M., and Bayru, A. (2015) One foot on the ground, one foot in the air. Ethiopia’s delivery on an ambitious development agenda. London: Overseas Development Institute.

Lewis, W.A. (1954) Economic development with unlimited supplies of labour. The Manchester School, 2, 139–191.

Ludi, E. (2004) Economic Analysis of Soil Conservation: Case studies from the Highlands of Amhara Region, Ethiopia. Berne: Centre for Development and Environment.

Makki, F. and Geisler, C. (2011) Development by dispossession: Land grabbing as new enclosures in contemporary Ethiopia. In International Conference on global land grabbing, vol. 68. 2011.

Mansour, D. (2016) The socio-economic of exclusion: Re-questioning citizenship in relation to social justice in the post Arab Spring Egypt. Arab Citizenship Review, 15, 19 February 2016.

Marcus, R., and Harper, C. (2014) Gender Justice and Social Norms - Processes of Change for Adolescent Girls. London: Overseas Development Institute.

Marcus, R., and Harper, C. (2015) How do gender norms change? London: Overseas Development Institute.

Mariotti, C., and Diwakar, V. (2016) Ensuring escapes from poverty are sustained in rural Ethiopia. Washington, DC: USAID.

Mayer, E. (2009) Cuentos feos de la reforma agraria Peruana. Lima: Institute de Estudio Peruanos.Mbilinyi, M. (1994) Restructuring gender and Agriculture in Tanzania. African perspectives on

development: Controversies, dilemmas and openings, 166-182.

153

McMillan, M., Page, J., and te Velde, D. (2015) Supporting economic transformation. London: Overseas Development Institute.

McMillan, M., Rodrik, D., Verduzco-Gallo, I., et al. (2014) Globalization, structural change, and productivity Ggowth, with an update on Africa. World Development, 63 (1), 11–32.

Mdoe, N. and Wiggins, S. (1997) Returns to smallholder dairying in the Kilimanjaro region, Tanzania. Agricultural Economics, 17, 75–87.

Mekonnen, A., and Asrese, K. (2014) Household decision making status of women in Dabat district, North West Ethiopia, 2009 Gc. Science Journal of Public Health, 2(2), 111–118.

Mellor, J., and Gavian, S. (1999) The Determinants of Employment Growth In Egypt—The Dominant Role of Agriculture And the Rural Small Scale Sector. Cambridge. MA: Abt Associates Inc. Mimeo.

Melmed‐Sanjak, J. S. and Carter, M.R. (1991) The economic viability and stability of ‘capitalised family farming’: An analysis of agricultural decollectivisation in Peru. Journal of Development Studies 27 (2), 190–210.

Mills, M. B. (2003) Labor migrants and marriage choices in contemporary Thailand. Family: Critical Concepts in Sociology, 4, 26.

Mills, M. B. (2005) Engendering discourses of displacement Contesting mobility and marginality in rural Thailand. Ethnography 6(3), 385-419.

Mills, M. B. (2016) Gendered Morality Tales: Discourses of gender, labour, and value in globalising Asia. The Journal of Development Studies, 1-15.

Mills, M. B. (2001). Rural-Urban Obfuscations: thinking about urban anthropology and labor migration in Thailand. City and Society 13(2), 177–182.

Minot, N., and Mekonnen, D. (2012) Role of agricultural cooperatives and storage in rural Ethiopia: Results of two surveys. Washington, DC: IFPRI.

MoFED. (2013) Development and Poverty in Ethiopia. Addis Abeba: Ministry of Finance and Economic Development.

Molle, F., Floch, P., Promphakping, B. and Blake, J. (2009) The’greening of Isaan’: Politics, ideology, and irrigation development in the Northeast of Thailand. Contested waterscapes in the Mekong region: Hydropower, livelihoods and governance: 253-282.

Molony, T. (2008) Running out of credit: the limitations of mobile telephony in a Tanzanian agricultural marketing system. Journal of Modern African Studies, 46 (4), 637–658.

Moore, B. (1966) The social origins of dictatorship and democracy. Lord and peasant in the making of the modern world. Harmondsworth, Middlesex: Penguin.

Mosley, P. (1985) Achievements and contradictions of the Peruvian agrarian reform: A regional perspective. Journal of Development Studies, 21 ( 3), 440–448.

Mujica, J. (2011) Violaciones sexuales en el Perú 2000-2009. Un informe sobre el estado de la situación. Lima: Promsex

Mulema, A., and Damtew, E. (2016) Gender-based constraints and opportunities to agricultural intensification in Ethiopia: A systematic review. Addis Abeba: ILRI.

Mutabazi, K., Wiggins, S. and Mdoe, N. (2013) Commercialisation of African Smallholder Farming. The Case of Smallholder Farmers in central Tanzania. Future Agricultures Working Paper 72. Brighton: Institute of Development Studies.

Naguib, N. G., and Lloyd, C. B. (1994) Gender inequalities and demographic behavior: the case of Egypt. [Unpublished] 1994. Presented at the Population Council Symposium on Family Gender and Population Policy: International Debates and Middle Eastern Realities Cairo Egypt February 7–9 1994.

National Statistical Office (2012) Thailand. Multiple Indicator Cluster Survey 2012, BangkokNational Statistical Office (2016) Thailand. Multiple Indicator Cluster Survey 2014/15, BangkokO’Sullivan, M. (2016) Thematic White Paper on Gender and Property Rights. World Bank Gender

Innovation Lab. Washington, DC: World Bank.

154

Oumer, A. M., Tiruneh, W. G., and Tizal, C. Y. (2014) Empowering smallholder women farmers through participatory seed potato management: Lessons from Welmera District, Ethiopia. Journal of Sustainable Development, 7(5), 93–110.

Oxfam (2017) Oxfam’s Conceptual Framework on Women’s Economic Empowerment. Oxford: Oxfam International.

Oxford Poverty and Human Development Initiative. (2013) Ethiopia Country Briefing. Oxford: University of Oxford.

Palacios-Lopez, A., Christiaensen, L., and Kilic, T. (2017) How much labour in Africa is provided by women? Food Policy, 52–63.

Parker, H., Mosello, B., Calow Roger, Quattri, M., Kebede, S., and Alamirew, T. (2016).A thirsty future? Water strategies for Ethiopia’s new development era. London: Overseas Development Institute.

Parker, H., Oates, N., Mason, N., Calow, R., Chadza, W., and Ludi, E. (2016) Gender, agriculture and water security. London: Overseas Development Institute.

Persha, L., Greif, A., and Huntignton, H. (2017) Assessing the impact of second-level land certification in Ethiopia. Washington, DC: 2017 World Bank Conference on Land and Poverty.

Persha, L., Greif, A. and Huntington, H. (2017) Assessing the Impact of Second-Level Land Certification in Ethiopia.” In Paper prepared for presentation at the 2017 World Bank Conference on Land and Poverty. Washington DC: The World Bank.

Plaza, O., and Stromquist, N. P. (2006) Consequences of Structural Adjustment on Economic and Social Domains: Two Decades in the Life of Peru. The ANNALS of the American Academy of Political and Social Science, 606(1), 95–115. https://doi.org/10.1177/0002716206288759

Ponte, S. (2001) Trapped in decline? Reassessing agrarian change and economic diversification on the Uluguru Mountains, Tanzania. Journal of Modern African Studies, 39 (1), 81–100

Promphakping, B. (2000) Rural Transformation and Gender Relations in the Northeast of Thailand. Bath: University of Bath.

Prüss-Üstün, A., Bos, R., Gore, F., and Bartram, J. (2008) Safer water, better health. Costs, benefits and sustainability of interventions to protect and promote health. Geneva: World Health Organisation.

Putterman, L. (1995) Economic reform and smallholder agriculture in Tanzania: a discussion of recent market liberalization, road rehabilitation, and technology dissemination efforts. World Development, 23 (2), 311–26.

Quisumbing, A., and Pandolfelli, L. (2009) Promising approches to address the needs of poor female farmers. Washington, DC: IFPRI.

Rani, D., and Yadeta, D. (2016) Empowerment of rural women through cooperative: A case of saving and credit cooprative society of Dendi district, Oromia regional state, Ethiopia. International Journal of Applied Research, 367–373.

Rashid, S. and Dorosh, P. (2008) The Interplay of Public Policy and Grain Market Behavior in Ethiopia. Briefing note prepared for the DFID funded study “Understanding the constraints to continued rapid growth in Ethiopia: the role of agriculture”. http://users.ox.ac.uk/~econstd/Brief_SRashid_v2.pdf.

Rashid, S. and Negassa, A. (2011) Policies and Performance of Ethiopian Cereal Markets. Ethiopia Strategy Support Program II (ESSP II). Working Paper No. 21, May 2011. Addis Ababa and Washington, DC: IFPRI.

Rasmussen, T. (1986) The Green Revolution in the Southern Highlands. In Boesen, J., K.J., Havnevik, Koponen, J., and Odgaard, R. (eds.) Tanzania. Crisis and struggle for survival. Uppsala: Scandinavian Institute of African Studies.

Rigg, J. (2019) More than rural: textures of Thailand’s agrarian transformation. Honolulu: Hawaii University Press.

155

Rigg, J., Salamanca, A. and Parnwell, M. (2012) Joining the Dots of Agrarian Change in Asia: A 25 Year View from Thailand. World Development 40 (7), 1469–1481, http://dx.doi.org/10.1016/j.worlddev.2012.03.001

Rigg, J., Promphaking, B. and Le Mare, A. (2014) Personalizing the Middle-Income Trap: An Inter-Generational Migrant View from Rural Thailand, World Development 59, 184–198, 2014, http://dx.doi.org/10.1016/j.worlddev.2014.01.031

Rock, M. T. (2002) Exploring the impact of selective interventions in agriculture on the growth of manufactures in Indonesia, Malaysia, and Thailand, Journal of International Development, 14, 485–510.

Rodrik, D. (2015) Premature Deindustrialization. Cambridge, MA: National Bureau of Economic Research. http://www.nber.org/papers/w20935.

Rosenfield, A.G. and Min, C.J. (2007) The Emergence of Thailand’s National Family Planning Program, in Robinson, W.C. and Ross, J.A. (eds) The Global Family Planning Revolution: Three Decades of Population Policies and Programs. Washington, DC: World Bank.

Rugumamu, C. P. (2009).Assessment of post-harvest technologies and gender relations in maize loss reduction in Pangawe village eastern Tanzania. Tanzania Journal of Science, 35(1), 67-76.

Salem, S., and Gleason, J. (2003) An Examination of Poverty Reduction in Egypt: Contributing Factors, Sustainability, and Lessons (Pro-Poor Economic Growth Research Studies No. PCE-I-02-00-00015-00). Development Alternatives Inc./Boston Institute for Developing Economies.

Samman, E., Presler-Marshall, E., and Jones, N. (2016) Women’s work. Mothers, children and the global childcare crisis. London: Overseas Development Institute.

Sandford, J., and Hobson, M. (2011) Leaving No-One Behind: Ethiopia’s Productive Safety Net and Household Asset Building Programmes. World Bank: Washington, DC, USA.

Schaefer, F., and Abebe, G. (2015) The case for industrial policy and its application in the Ethiopian cut flower sector (No. 012).

Schmidt, E., and Bekele, F. (2016) Rural youth and employment in Ethiopia. Ethiopian Development Research Institute, Ethiopia Strategy Support Program. Washington, DC: International Food Policy Research Institute.

Schuster, M. and Maertens, M. (2015) Do labor standards benefit workers in horticultural export chains in Peru? Paper, Interrnational Conference of Agricultural Economist, Milan August 2015.

Semahegn, A., and Mengistie, B. (2015) Domestic violence against women and associated factors in Ethiopia; systematic review. Reproductive health, 12(1), 78.

Shiferaw, A., Söderbom , M., Siba, E., and Alemu, G. (2012) Road infrastructure and enterprise development in Ethiopia. London: International Growth Centre.

Sieverding, M., and Elbadawy, A. (2016) Empowering Adolescent Girls in Socially Conservative Settings: Impacts and Lessons Learned from the Ishraq Program in Rural Upper Egypt. Studies in Family Planning, 47(2), 129–144. https://doi.org/10.1111/j.1728-4465.2016.00061.x

Smith, N. M. (2015) Gender and livelihood diversification: Maasai women’s market activities in Northern Tanzania. The Journal of Development Studies, 51(3), 305-318.

Snyder, K., and Cullen, B. (2014) Implications of sustainable agricultural intensification for family farming in Africa: Anthropological perspectives. Anthropological Notebooks, 9–29.

Sonbol, Amira El-Azhary (2005) Egypt. Pages 69-86 in: Nazzir and Tomppert (eds). Women’s Rights in the Middle East and North Africa. Freedom House, Oxford. Available online at: http://www.refworld.org/docid/47387b6a46.html

Tafere, Y., and Pankhurst, A. (2015) Can children in Ethiopian communities combine schooling with work? Oxford: Young Lives, University of Oxford.

Taylor, J. E. and Adelman, I. (1996) Village economies. The design, estimation, and use of villagewide economic models. Cambridge: Cambridge University Press.

Tenaw, S., Islam, K., and Parviainen, T. (2009) Effects of land tenure and property rights on agricultural productivity in Ethiopia, Namibia and Bangladesh. Helsinki: University of Helsinki.

156

Tesfaye, A. (2015) Access of rural women to agricultural extension services. Opportunities and challenges in Ethiopia. Addis Abeba: Women and Youth Affaits Directorate of the Ministry of Agriculture and Agricultural Transformation Agency.

Timmer, C. P. and Akkus, S. (2008) The Structural Transformation as a Pathway out of Poverty: Analytics, Empirics and Politics. Center for Global Development Working Paper, no. 150. Washington, DC: Center for Global Development. http://papers.ssrn.com/sol3/papers.cfm?abstract_id=1213154.

Timmer, C. P. (2009) A World without Agriculture. The Structural Transformation in Historical Perspective. Washington, DC: AEI Press Publisher for the American Enterprise Institute.

Timmer, M. P., de Vries, G. J., and de Vries, K. (2015) Patterns of Structural Change in Developing Countries. In Weiss, J. and Tribe, M. (eds). Routledge Handbook of Industry and Development. Abingdon: Routledge.

Timmer, P. (1988) The Agricultural Transformation. In Chennery, H. and T.N. Srinivasan (eds). Handbook of Development Economics, Volume I. Elsevier Science Publishers B.V.

Timmer, P., McMillan, M., Badiane, O., Rodrik, D., Binswanger-Mkhize, H., and Wouterse, F. (2012) Patterns of Growth and Structural Transformation in Africa. http://pages.au.int/sites/default/files/IFPRI%20WCAO%20TRN2%20Structural%20Change_0.pdf.

Toth, J. (1991) Pride, Purdah, or Paychecks: What Maintains the Gender Division of Labor in Rural Egypt? International Journal of Middle East Studies, 23(2), 213–236. https://doi.org/10.1017/S0020743800056051

Toth, J. (2002) Rural-to-Urban Migration and Informal Sector Expansion: Impediments to Egyptian Development, Robert Schuman Centre for Advanced Studies (RSC) No. 2002/15, Mediterranean Programme Series, European University Institute WORKING PAPERS, San Domenico (FI), Italy

UN Habitat (2008) Land Certification in Ethiopia - Early Impacts on Women. Nairobi: United Nations Human Settlement Programme.

UN Women (2014) Preliminary Gender Profile of Ethiopia. Addis Abeba: UN Women.UNDP (2015) Accelerating Inclusive Growth for Sustainable Human Development in Ethiopia.

National Human Development Report 2014. Addis Abeba: United Nations Development Programme.

Unicef and World Health Organisation (2015) Progress on sanitation and drinking water – 2015 update and MDG assessment. New York and Geneva: Unicef and World Health Organisation.

Unicef (2013) Female Genital Mutilation / Cutting: A statistical overview and exploration of the dynamics of change. New York: Unicef.

UNICEF (2016) Country statistics. Available online at: https://www.unicef.org/infobycountry/egypt_statistics.html

UNICEF and WHO (2015) 25 Years of Progress on Sanitation and Drinking Water. 2015 Update and MDG Assessment. UNICEF, New York and WHO, Geneva.

USAID (2011) USAID/Ethiopia Feed the Future Gender Analysis. Addis Ababa: Kansas State University.

USAID (2016) Fact Sheet: Land Tenure and Women’s Empowerment. Washington, DC: USAID.Warner, J., Kieran, C., and McMullan, S. (2015) A review of selected topics of gender and agricultural

research in Ethiopia over the last decade. Washington, DC and Addis Abeba: IFPRI.Watson, C., and Harper, C. (2016) How communications can change social norms around adolescent

girls. Lessons learned from year 3 of a multi-country field study. London: Overseas Development Institute.

WHO (2004) Thailand and family planning: An overview. Brief. New Delhi: Regional Office for South-East Asia, World Health Organisation.

WHO (2005) WHO multi-country study on women’s health and domestic violence against women. Initial results on prevalence, health outcomes and women’s responses. Geneva: WorldHealth Organisation.

157

Wiggins, S. (2014) Rural non-farm economy: current understandings, policy options, and future possibilities. In Hazell, P. B. R. and Rahman, A. (eds). New directions for smallholder agriculture. Oxford: Oxford University Press.

Wiggins, S. and Keats, S. (2014) Rural wages in Asia. London: Overseas Development Institute. Wongmonta, S. and Glewwe, P. (2016) An analysis of gender differences in household education

expenditure: the case of Thailand. Education Economics, 1-22.Woodhouse, P., Veldwisch, J.V., Venot, J.-P., Brockington, D., Komakech, H., and Manjichi, A. (2017)

African farmer-led irrigation development: re-framing agricultural policy and investment? The Journal of Peasant Studies, 44 ( 1), 213–233.

World Bank and ONE (2014) Levelling the field. Improving opportunities for women farmers in Africa. Washington DC: The World Bank.

World Bank (2006) World development report 2006: Equity and development. New York: Oxford University Press.

World Bank (2010) The Ethiopian Urban Migration Study 2008: The characteristics, motives and outcomes of migrants to Addis Abeba. Addis Abeba: World Bank.

World Bank (2012) Agribusiness Indicators: Ethiopia. Washington, DC: World Bank.World Bank (2013) Ethiopia’s Productive Safety Net Programme (PSNP). Integrating Disaster and

Climate Risk Management. Washington, DC: World Bank.World Bank (2014) Ethiopia’s Poverty Assessment. Poverty Global Practice Africa Region. Addis

Ababa: World Bank Group.World Bank. (2015a) Ethiopia Poverty Assessment. Washington, DC: World Bank.World Bank (2015b) Global Monitoring Report 2014/15: Ending Poverty and Sharing Prosperity.

Washington, DC: World Bank.World Bank (2015c) 4th Ethiopia Economic Update. Overcoming constraints in the manufacturing

sector. Washington, DC: World Bank.World Bank (2016) Ethiopia’s Great Run. The growth acceleration and how to pace it. Washington,

DC: World Bank.World Bank (2017) Doing Business 2017: Equal Opportunities for All. Washington, DC: World Bank.World Bank, Cities Alliance (2015) Ethiopia Urbanisation Review. Urban Institutions for a middle-

income Ethiopia. Washington, DC: World BAnk.Yamada, G. (2010) Growth, Employment and Internal Migration. Peru, 2003-2007, Working Paper,

Centro de Investigación de la Universidad del Pacifico, Lima. Online at https://mpra.ub.uni-muenchen.de/22067/

Yoder, P., and Wang, S. (2013) Female genital cutting: the interpretation of recent DHS data. Calverton: ICF International.

Zaldívar, R. (1974) Agrarian reform and military reformism in Peru. In Lehmann, D. (ed.) Agrarian Reform and Agrarian Reformism. Faber and Faber.

Zerihun, A., Wakiaga, J., and Kibret, H. (2016) Ethiopia 2016. African Economic Outlook. AfDB, OECD, UNDP.

Zohry, A. (2005) Interrelationships between Internal and International Migration in Egypt: A Pilot Study. Paper, Forced Migration and Refugee Studies Program, American University in Cairo and Development Research Centre on Migration, Globalisation and Poverty, University of Sussex.

Evidence.Ideas.Change.

ODI203 Blackfriars RoadLondon SE1 8NJ

+44 (0)20 7922 0300info@odi.org

odi.orgodi.org/facebookodi.org/twitter

ODI is an independent, global think tank, working for a sustainable and peaceful world in which every person thrives. We harness the power of evidence and ideas through research and partnership to confront challenges, develop solutions, and create change.